chore(release): 1.4.0

`tags: [a, b, c]` is the form SPEC.md 4.1's own frontmatter skeleton writes out,
and 9/53 upstream reference concepts use it. It raised on the `[` indicator.
It parses now, to the same list its block-sequence sibling already produced.

The predicate is character-level, inside `_parse_flow_sequence`: an element is a
plain scalar only if it is non-empty and carries none of `{ } [ ] : , " ' #`,
and it then passes the unchanged scalar-indicator rule. Everything that would
need YAML semantics to split or unquote still raises - a quoted element (quotes
are retained here, never stripped), a colon or comma inside an element, a
sequence inside a sequence, an empty element, an anchor, an alias. A sequence
may not mix scalars and mappings, the rule the block list already carries, and
the mixing verdict is reached before the element is parsed so the caller is told
about the mix rather than about a key the allowlist would have named instead.
The 1.3.0 `sources` flow-mapping carrier is unchanged and pinned against
regression. Depth 1 is not spent: the elements are leaves.

Measured with the denominator, against the pinned corpus (`_okf-upstream` @
3fcbb9f, 53 non-reserved documents) and the pinned SPEC (`_okf-canonical` @
ad30107). Baseline reproduced first, with a known-positive control, at 0/53.
After: 6/53, all six in acme_retail. It does not close the corpus - 44/53 still
stop on `generated` written as a top-level block mapping, which spends the
no-nesting-past-depth-1 rule and is a security decision, out of scope here.

P1 alone, per the operator decision of 08.09. The two neighbouring predicates
were measured and deliberately not built: a flush-left block sequence and a
folded plain scalar release 0/53 each on their own, and stacked on this one they
still measure 6/53. `_consume_block_list`, the `description` continuation and
the allowlist are untouched.

docs/LIMITATIONS.md's tags/description entry is rewritten against the
measurement: three of its claims were wrong. The parser does have a
sequence-value type (since 1.3.0 - what it lacks is the indentation the corpus
omits); the figure is 6/53, not the 1.2.0-era 4/53; and tags/description are
not the residual that blocks the corpus. README gains the sequence carrier in
the paragraph that already describes the mapping one.

Self-safety: the predicate compiles no regex, so docs/redos-sweep.py cannot see
it. Measured instead on the CPU clock - linear in element length (exponent
0.86-0.99) and in element count (0.97-1.05) over four doublings to 800_000 -
and pinned by two bounds in tests/test_okf.py.

Six rows that pinned the old refusal are re-aimed at the class that still
holds - the quoted element - the way the 1.3.0 rows were when the carrier
opened. One of them lives in src/llm_ingestion_guard/coverage.py, which is why
the src diff is three files rather than one.

Version 1.4.0 in the code only. The CHANGELOG entry stays under Unreleased and
no tag is cut: README's badge and install pin must keep naming a tag that
exists.

Gates after `git add`: 893 passed (was 868), coverage 130/130 + 6/6 gaps,
redos-sweep exit 0, LIMITATIONS still 45 entries.

.gitignore

@@ -24,3 +24,6 @@ coverage/
 *.local.md
 *.local.sh
 .DS_Store
+
+# --- scratchpad: measurement scripts are tracked, the extracted corpus copy is not ---
+/scratchpad/corpus/

CHANGELOG.md

@@ -5,9 +5,166 @@ All notable changes to this project will be documented in this file.
 The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.1.0/),
 and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html).
 
-## [Unreleased]
+## [1.4.0] — 2026-09-08
 
-Nothing yet.
+### Added — `tags: [a, b, c]`, the one flow-sequence form SPEC §4.1 writes out
+
+`okf.parse_frontmatter` refused a flow sequence of plain scalars. That is the
+form SPEC.md §4.1's own frontmatter skeleton uses for `tags`, and `tags` is
+present in 53/53 upstream reference concepts, 9/53 of them in exactly this form.
+Admitted now, parsing to the same value its block-sequence sibling already did:
+
+```yaml
+tags: [finance, revenue, headline-metric]
+```
+
+An element is a plain scalar only if it is non-empty and carries none of
+`{ } [ ] : , " ' #`, and it then passes the unchanged scalar-indicator rule.
+The refusals that make that a rule rather than a preference: a quoted element
+(this parser retains quotes rather than stripping them, so reading one would
+hand back a value YAML does not), an element carrying a colon or a comma, a
+sequence inside a sequence, an empty element, an anchor or an alias. A sequence
+may not **mix** scalars and mappings — the same rule the block list already
+carries, for the same reason: a consumer iterating the value and reading
+`entry.get("id")` crashes on the first `str`. The `sources` flow-mapping
+carrier added in 1.3.0 is unchanged and pinned against regression.
+
+**Measured, with the denominator: this takes the pinned OKF corpus from 0/53 to
+6/53** (`_okf-upstream/okf` @ `3fcbb9f`, 53 non-reserved documents; the six are
+all in `acme_retail`). It does not close the corpus. **44/53 still stop on
+`generated` written as a top-level block mapping** — a form refused on the
+no-nesting-past-depth-1 rule, which is a security decision and out of scope
+here. Two neighbouring predicates were measured and deliberately not built: a
+flush-left block sequence and a folded plain scalar release **0/53** each on
+their own and leave the corpus at 6/53 when stacked on this one. Detection
+behaviour is not frozen under semver; the exported surface is unchanged.
+
+### Changed
+
+- `docs/LIMITATIONS.md`'s `tags`/`description` entry is rewritten against the
+  measurement. Three of its claims were wrong: the parser *does* have a
+  sequence-value type (since 1.3.0 — what it lacks is the indentation the
+  corpus omits), the figure is 6/53 and not the 1.2.0-era 4/53, and
+  `tags`/`description` are *not* the residual that blocks the corpus.
+- Self-safety: the new predicate is character-level with no regex, so
+  `docs/redos-sweep.py` cannot see it (it collects compiled patterns). It is
+  measured instead — linear in both element length (exponent 0.86-0.99) and
+  element count (0.97-1.05) over four doublings to 800_000, and pinned by two
+  CPU-clock bounds in `tests/test_okf.py`.
+
+## [1.3.0] — 2026-09-02
+
+### Added — the `sources` provenance family becomes expressible in both spec carriers
+
+`okf.parse_frontmatter` rejected `sources` in every form the OKF spec and its
+producers actually use. Measured 2026-09-02 by two consumers independently:
+`sources: [{ id: a, resource: x }]` raised on the `[` indicator (with one entry
+as with two), and the block sequence of block mappings — SPEC.md §5.1's *own*
+example — raised `"nested mappings are not supported"`. `resource` is REQUIRED
+within a `sources` entry (§5.1), so the whole provenance family was unwritable,
+and a bundle written the way the spec documents it was refused.
+
+Admitted now, both parsing to the same value (a list of dicts):
+
+```yaml
+sources: [{ id: a, resource: https://e.com/a }]
+sources:
+  - id: a
+    resource: https://e.com/a
+```
+
+**`resource` is allowlisted inside a `sources` entry and nowhere else.** 1.2.0
+left it off the allowlist on the argument that the parser could not tell
+`sources[].resource` (§5.1, a citation) from `executor.resource` /
+`attester.resource` (§10, a pointer to code to be run — the door-C route closed
+in 1.1.0). That premise was false: the owning key is in scope at every call
+site and was simply never threaded through. It is threaded now, so the
+discrimination is structural rather than a judgement about the value, and
+`executor: [{ resource: skills/run.md }]` and `attester:\n  - resource: …` are
+refused on the allowlist through *every* carrier, including the two this adds.
+
+Refusal stays the default everywhere else. A flow sequence of plain scalars
+(`tags: [a, b]`) still raises — the sequence carrier is opened for the flow
+mapping element and nothing else. A `sources` entry still admits scalar leaves
+only, so SPEC §5.1's optional *per-entry* `usage_window` (a mapping inside a
+mapping) is refused: no nesting past depth 1 is a security property, and it was
+not spent here. Registered as a conformance gap in `docs/LIMITATIONS.md`. A
+block list may not mix scalar items and mappings. Off-allowlist keys, anchors,
+aliases, tags, duplicate keys and unclosed collections raise as before, and a
+refused mapping still raises rather than degrading into a string.
+
+**New residual, registered:** `sources[].resource` is scanned as text (T1) but
+never validated as a URL. T3's https allowlist cannot be extended to it without
+over-blocking conformant bundles — §5.1 permits a bundle-relative path or a
+scope descriptor a consumer cannot follow at all. A consumer that dereferences
+it must call `okf.validate_resource_url` itself.
+
+No exported surface changed and no detector behaviour or calibration changed.
+Suite 834 → 868 (34 new rows, plus four pre-existing rows updated where this
+release changed the behaviour they pinned); 130/130 classes, 6/6 documented
+gaps hold, 45 limitations, ReDoS sweep 0/152 candidates flagged.
+
+## [1.2.0] — 2026-08-23
+
+### Added — OKF frontmatter can express one mapping form: typed and allowlisted
+
+`okf.parse_frontmatter` gave the mapping *class* no expressible form at all. OKF
+v0.2 writes its whole trust and provenance layer as mappings — SPEC.md @
+`62432a09` uses flow form in its own §5.1/§5.2 examples, and §11 carries a hard
+MUST for consumers ("MUST treat a bare `verified` mapping as a one-element
+list") that presupposes they parse. A consumer measured **0 of 53** upstream
+concepts through the gate on 0.3.4, 1.0.0 and 1.1.0. That was a contract
+collision, not a calibration setting: no threshold would have moved it.
+
+Admitted now: a flow mapping (`generated: { by: x, at: y }`), as a value or as a
+block-list item, whose every key is on a nine-name allowlist (`by`, `at`, `from`,
+`to`, `id`, `title`, `author`, `usage_count`, `last_modified`) and whose every
+leaf is a plain scalar run through the *unchanged* dangerous-value and
+mapping-construct predicates.
+
+The form is additive and refusal stays the default. A key off the allowlist, a
+nested collection, a quoted leaf, a duplicate key, an empty or unclosed mapping,
+and `{a:b}` (which PyYAML 6.0.3 reads as the *key* `a:b`) all raise, and a
+refused mapping still raises rather than degrading into a string — the 1.1.0
+defect is not reopened. Nested-block (`k:\n  sub: v`), dotted (`k.sub: v`) and
+inline-second-colon (`k: sub: v`) routes to a mapping still raise, each on its
+own rule.
+
+**`resource` is deliberately off the allowlist**, though SPEC.md §5.1 names it
+inside a `sources` entry. It is a pointer rather than a label and the only key
+T3 exists for: admitting it would let `executor: { resource: skills/run.md }`
+carry an executable-code pointer through in typed clothes, which is the door-C
+route closed in 1.1.0. It costs nothing today — the conformant carrier for
+`sources[].resource` is a block sequence of block mappings, which this form does
+not admit either way.
+
+Mapping leaves are scanned like every other frontmatter value (T1), so an
+injection parked in `generated: { by: ... }` reaches `scan_output`. Coverage
+matrix: 130 classes, up from 129 (the new row is the off-allowlist key).
+
+No exported surface changed; no detector behaviour and no calibration changed.
+
+
+### Changed — the ReDoS sweep now measures on the same clock as the bounds it justifies
+
+`docs/redos-sweep.py` timed on `time.monotonic()` while every ReDoS bound in the
+suite moved to process CPU time (`tests/redos_clock.py`), so the 1.5 ms
+sensitivity floor and the "~23 s at the cap" figure published in
+`docs/LIMITATIONS.md` came from a different instrument than the bounds they
+support. The script now imports `scan_seconds` rather than timing itself.
+
+The floor was re-derived on that instrument and **stayed at 1.5 ms**: over twelve
+full runs of all 2585 arms the median ratio is 1.95-2.03 in every size bucket
+above 50 µs, but two-point excursions past the 2.6 flag threshold persist at every
+magnitude (p99 ratio 2.9-3.3 even above 1 ms) — 6.9 flagged arms per run at a
+0.5 ms floor, 1.1 at 1.0 ms, 0.33 at 1.5 ms. Descheduling was never what made this
+sweep noisy; a ratio computed from two points is. Four arms flagged across those
+twelve runs, each in exactly one of them, and six arms that have ever flagged
+re-measure at exponent 0.97-1.09 over six doublings — at most 1.2 s at the
+1 000 000-char cap. The pattern count the script prints is 152, not the 150 of the
+0.3.4 entry below; `docs/LIMITATIONS.md` now carries the current number.
+
+No exported surface, no detector behaviour and no calibration changed.
 
 
 ## [1.1.0] — 2026-08-13
@@ -33,6 +190,12 @@ while closing it**, and is the reason this release names two routes rather than
 Neither shape is conformant OKF — a well-formed bundle does not produce them; a
 malformed or hostile one can.
 
+**What closed is the type confusion, not pointer-smuggling as a class.** T3 still
+inspects `resource` and nothing else, so an honest string under another key rides
+through exactly as before: `attester: attesters/sql_equality.py` is WARN, while
+the same path under `resource:` FAIL_SECUREs. The string is still scanned like any
+other frontmatter value under T1. Nothing about that changed here.
+
 **The boundary is where YAML puts it**, ground-truthed against PyYAML 6.0.3 rather
 than reasoned: `": "` and a trailing `":"` are exactly the two shapes where a plain
 scalar becomes a mapping, and they are refused. A colon carrying neither a space nor
@@ -46,8 +209,10 @@ here — the `1.0.0` entry says exactly this is a fix, not a break. A consumer w
 bundles carry an unquoted `": "` in a frontmatter value will now see those concepts
 refused at import; quote the value, and it parses.
 
-Suite 792 → **802** (11 rows added, 1 retired). 129/129 classes, 6/6 documented
-gaps, 35 limitations — all unchanged.
+Suite 792 → **802**: 13 rows added (4 rejected shapes, 7 admitted ones, 2 through
+`import_bundle`), 3 retired (the two that pinned the defect, and the one-key row
+in the block-list table). 129/129 classes, 6/6 documented gaps, 35 limitations —
+all unchanged.
 
 
 ## [1.0.0] — 2026-08-13

CLAUDE.md

@@ -11,11 +11,40 @@ framework-agnostisk kode.
 Referanse-implementasjon: `claude-code-llm-wiki` Stage B (`tools/wiki_ingest/`).
 Lexikon-seed: `injection-patterns.mjs` fra `llm-security`-pluginen.
 
-Repoet er på **v1.1.0** — den eksporterte Python-surfacen er frosset under semver
+Repoet er på **v1.4.0 i koden, UUTGITT** (`pyproject.toml` + `__init__.py` er
+bumpet; README-badge, install-pinnen, ADOPTION-BRIEF og BRIEF står med vilje
+igjen på `1.3.0`, som er den siste taggen som FINNES — en install-pin må peke på
+en ekte tag). Release-commiten (CHANGELOG-overskrift datert, de fire
+dokumentflatene bumpet, tag) er ikke tatt. Den eksporterte Python-surfacen er frosset under semver
 (deteksjonsatferd er det IKKE; kalibrering flytter seg i 1.x). Stdlib-kjernen er
 bygget og testet (15 moduler +
 topp-nivå wiring, showcase + korpus), inkl. OKF-adapter og aktivt-innhold-
-detektor (EchoLeak-klassen) i output-gaten. Mode-b `import_bundle` skanner
+detektor (EchoLeak-klassen) i output-gaten. OKF-frontmatterens mapping-klasse
+har **fire** uttrykkbare bærere (G3 21.08, G30 02.09): flow-mapping som verdi
+og som blokkliste-element, flow-sekvens av flow-mappinger, og blokk-sekvens av
+blokk-mappinger (SPEC §5.1s egen form). HVER nøkkel i alle fire står på
+allowlisten og hvert blad er en ren skalar. Formen er trygg fordi allowlisten
+inspiserer hver nøkkel; det blanke avslaget var håndhevelsen, ikke poenget.
+**`resource` er allowlistet KUN inne i en `sources`-oppføring** — foreldre-
+nøkkelen avgjør, så `executor`/`attester` sin `resource` (§10, dør C) avvises
+gjennom hver eneste bærer. 1.2.0s begrunnelse for å utelate den (parseren
+manglet foreldre-kontekst) var målt feil: konteksten var der, den var bare
+aldri tredd gjennom. Topp-nivå blokk-mapping, dotted- og inline-kolon-rutene
+raiser fortsatt, en blokkliste kan ikke blande skalarer og mappinger, og en
+avvist mapping raiser — den degraderer aldri til en streng (1.1.0-defekten).
+**Flow-sekvens av rene skalarer (`tags: [a, b]`) PARSER fra 1.4.0** (P1,
+operatørbeslutning 08.09) — SPEC §4.1s eget skjelett. Et element er en ren
+skalar kun hvis det er ikke-tomt og uten `{ } [ ] : , " ' #`, og så gjelder den
+uendrete indikator-regelen; sitert element, kolon/komma i elementet, sekvens i
+sekvens, tomt element, anker og alias raiser fortsatt, og en flow-sekvens kan
+ikke blande skalarer og mappinger. **Målt med nevner: 0/53 → 6/53** på pinnet
+OKF-korpus (`3fcbb9f`). Den bindende skranken er IKKE tags/description, men
+`generated` som topp-nivå blokk-mapping (44/53) — den bruker opp dybde-1 og er
+en sikkerhetsbeslutning. P2 (blokksekvens uten innrykk) og P3 (foldet plain
+scalar) er MÅLT til 0/53 hver og bevisst IKKE bygget. `sources[].resource`
+URL-valideres ALDRI (T3 ser kun topp-nivå `resource`) — §5.1 tillater
+bundle-relative stier og scope-beskrivelser, så en https-gate ville over-blokkert
+konforme bundles; konsumenten må selv kalle `validate_resource_url`. Mode-b `import_bundle` skanner
 reserverte strukturfiler (`index.md`/`log.md`) i mottatte bundles i stedet for å
 path-avvise dem; upload-front-end beholder shadow-reject (`allow_reserved=False`).
 Output-gatens decode-and-rescan mater dekodet base64-klartekst gjennom BÅDE lexicon

README.md

@@ -2,7 +2,7 @@
 
 Write-time defensive layer for Python pipelines that persist LLM output: sanitize, fence, tool-less quarantined transform, capability isolation, scan before persist, fail-secure.
 
-![Version](https://img.shields.io/badge/version-1.1.0-blue)
+![Version](https://img.shields.io/badge/version-1.3.0-blue)
 ![Status](https://img.shields.io/badge/status-stable-brightgreen)
 ![Python](https://img.shields.io/badge/python-3.10%2B-purple)
 ![License](https://img.shields.io/badge/license-MIT-lightgrey)
@@ -33,7 +33,7 @@ at write time, never assumed from the format. Any pipeline ingesting external da
 into an agent-read store has this shape; an OKF wiki is its canonical form — which
 is why the guard ships a first-class OKF adapter (below).
 
-**Status:** `v1.1.0`. The stdlib-only core — its detector, contract, and
+**Status:** `v1.3.0`. The stdlib-only core — its detector, contract, and
 OKF-adapter modules plus the top-level wiring — is built and tested, exercised by
 an end-to-end showcase and adversarial + false-positive corpora. The exported
 Python surface is now frozen under semver: nothing exported is removed, renamed or
@@ -41,12 +41,24 @@ given a different meaning without a `2.0.0`. **Detection behaviour is not frozen
 severities, thresholds and lexicon entries are calibration and move in `1.x`. There
 are real limitations, stated plainly below; read them.
 
+## Table of Contents
+
+- [Install](#install)
+- [Quickstart — the two bookends](#quickstart--the-two-bookends)
+- [OKF / LLM-wiki support (shipped)](#okf--llm-wiki-support-shipped)
+- [What it protects against](#what-it-protects-against)
+- [The reusable contract (adopt-this checklist)](#the-reusable-contract-adopt-this-checklist)
+- [Known limitations](#known-limitations)
+- [Non-goals](#non-goals)
+- [Design & threat model](#design--threat-model)
+- [License](#license)
+
 ## Install
 
 Not on PyPI. The guard is distributed from its Forgejo origin — pin a release tag:
 
 ```bash
-pip install "llm-ingestion-guard @ git+https://git.fromaitochitta.com/open/llm-ingestion-pipeline-security.git@v1.1.0"
+pip install "llm-ingestion-guard @ git+https://git.fromaitochitta.com/open/llm-ingestion-pipeline-security.git@v1.3.0"
 ```
 
 The `open/` mirror is anonymously readable, so CI needs no deploy key, token, or
@@ -153,7 +165,19 @@ Per-concept gates: **path / reserved-name** (rejects `..` traversal and reserved
 reject-by-default loader that refuses anchors, aliases, and explicit tags *by
 construction*, so a billion-laughs alias expansion or a `!!python/object` coercion
 cannot occur (it is deliberately **not** a general YAML engine, whose own features
-are the attack surface); **`resource` https-allowlist** (hard-rejects
+are the attack surface). The one mapping form it accepts is OKF v0.2's flow
+mapping — `generated: { by: x, at: y }`, `verified: { … }` bare or listed,
+`usage_window: { from: …, to: … }` — admitted key-by-key against a nine-name
+allowlist (`by`, `at`, `from`, `to`, `id`, `title`, `author`, `usage_count`,
+`last_modified`) with plain-scalar leaves only. A key off that list, a nested
+collection or a duplicate key is refused, and `resource` is deliberately not on
+it; the block, dotted and inline-colon routes to a mapping still raise. A
+*sequence* value has two carriers — the block list, and (as of `1.4.0`) the flow
+sequence `tags: [a, b, c]`, which is SPEC §4.1's own skeleton — whose elements
+are either all plain scalars or all flow mappings, never a mix. A scalar element
+carrying any of `{ } [ ] : , " ' #` is refused rather than guessed at. See
+[LIMITATIONS](docs/LIMITATIONS.md) for what that admits and what it still walls
+off (a `sources` block list of mappings is still refused); **`resource` https-allowlist** (hard-rejects
 `data:`/`javascript:`/`file:` before commit — a reject-gate, not defang);
 **whole-concept scan** (frontmatter *values* + body through `scan_output`);
 **cross-link graph** (surfaces dangling targets, the dormant-injection signal, and
@@ -167,7 +191,7 @@ driven by a **live payload** in the coverage matrix — run it to watch all 134
 in your own environment:
 
 ```bash
-python -m llm_ingestion_guard.coverage    # 129/129 classes; exit 0 = all as documented
+python -m llm_ingestion_guard.coverage    # 130/130 classes; exit 0 = all as documented
 ```
 
 | Anchor | Attack classes it stops (representative) |
@@ -249,7 +273,7 @@ a green scan means safe content. The highest-impact items:
   two of the three corpora are living, so the cells are not rewritten in place.
   Method and before/after: [`docs/rawhtml-census.py`](docs/rawhtml-census.py).
 
-**Full list — 35 items, each with the mechanism, plus the out-of-scope boundary:**
+**Full list — 45 items, each with the mechanism, plus the out-of-scope boundary:**
 [`docs/LIMITATIONS.md`](docs/LIMITATIONS.md). Several carry field measurements from
 consumer corpora, including the false positives the URL-shape rule actually produces.
 

SECURITY.md

@@ -25,12 +25,12 @@ you happened to observe.
 **Do not open a public issue for a vulnerability.** Public disclosure before a fix
 gives an attacker a window against every downstream consumer.
 
-Instead, report it **privately** to the maintainer via the canonical repository on
-Forgejo:
+Instead, report it privately to <security@fromaitochitta.com> — mark the subject
+`SECURITY`.
 
-- Repository: `git.fromaitochitta.com/open/llm-ingestion-pipeline-security`
-- Contact the maintainer directly through that Forgejo instance (private message /
-  maintainer contact) and mark the subject `SECURITY`.
+- Canonical repository: https://git.fromaitochitta.com/open/llm-ingestion-pipeline-security
+- Alternatively, contact the maintainer directly through that Forgejo instance
+  (private message / maintainer contact) and mark the subject `SECURITY`.
 
 Please include:
 

docs/2026-09-07-limitations-44-maaling.md

@@ -0,0 +1,317 @@
+# Beslutningsgrunnlag — LIMITATIONS «punkt 44» (tags/description-gapet)
+
+**Målt:** 2026-09-07, mot `HEAD` = `44e2b31` (v1.3.0, urørt) og pinnet SPEC
+`_okf-canonical` @ `ad30107`. **Korpus:** `_okf-upstream` @ `3fcbb9f`.
+**Hva dette er:** underlaget for én operatørbeslutning — hvilket predikat, om
+noe, skal slippe inn den formen `tags` faktisk har i korpuset.
+**Hva dette ikke er:** beslutningen. Ingen fil i `src/` er endret, ingen
+versjon er bumpet, ingenting er pushet (0 upushede commits ved øktstart).
+
+---
+
+## 1. Referenten: «punkt 44» er ikke grepbart
+
+Ordren og STATE peker på «LIMITATIONS punkt 44». Målt:
+
+```
+$ grep -c '^- \*\*' docs/LIMITATIONS.md
+45
+$ grep -n '^- \*\*' docs/LIMITATIONS.md | awk -F: 'NR==44 {print $1}'
+756          # -> ZWJ-oppføringen, ikke tags/description
+```
+
+Tallet stammer fra commit `0184df9` (25.08), som skrev «43 -> 44 items» i
+commit-meldingen: **44 var antallet oppføringer etter innsettingen, ikke
+oppføringens posisjon.** Oppføringen ble satt inn midt i dokumentet og er i dag
+den **12. av 45**, på `docs/LIMITATIONS.md:126`. Etiketten «punkt 44» løser seg
+altså ikke opp verken ved posisjon eller ved `grep`, og den vil peke feil igjen
+neste gang listen vokser. Referenten er entydig i prosa («tags/description»),
+men nummeret bør ikke brukes videre.
+
+## 2. Hva punkt 44 sier i dag, ordrett
+
+`docs/LIMITATIONS.md:126-144`:
+
+> **`tags` and `description` block the OKF import corpus universally, before the
+> trust layer is even reached.** The line-flat frontmatter parser has no
+> sequence-value type at all: `tags` is present in 53/53 upstream concept
+> documents — 9/53 as a flow sequence (`[a, b, c]`, rejected on the `[`
+> indicator) and 44/53 as a block sequence (`- a` / `- b`, rejected as
+> `"malformed frontmatter line"`) — 100% rejection regardless of form.
+> `description` is present in 53/53; 29/53 is a folded plain scalar continuing
+> on an indented second line, which the parser has no continuation-line model
+> for and misreads as `"nested mappings are not supported"` (the remaining
+> 24/53 are single-line and parse fine). Measured directly on the upstream
+> reference bundles (`_okf-upstream/okf` @ `3fcbb9f`): removing `tags` alone
+> lets 4/53 documents pass; removing both `tags` and `description` together
+> (trust layer untouched) lets the same 4/53 pass, and all four then parse
+> `generated` correctly as a mapping. **Independent of the mapping-form work
+> above:** neither `1.2.0`'s flow mapping nor `1.3.0`'s `sources` carriers move
+> anything on this corpus, because `tags`/`description` reject before `sources`
+> is ever read. No sequence-value type or continuation-line model exists in the
+> stdlib-only parser to close this with.
+
+## 3. SPEC, ordrett fra pinnet kopi (`ad30107`)
+
+**§2 Terminology** (`SPEC.md:81`):
+
+> - **Frontmatter**: A YAML metadata block delimited by `---` at the top of
+>   a markdown file.
+
+**§4 Concept documents** (`SPEC.md:157-159`):
+
+> 1. A **YAML frontmatter block**, delimited by `---` on its own line at the
+>    start of the file and a closing `---` on its own line.
+
+**§4.1 Frontmatter**, skjelettet (`SPEC.md:163-173`):
+
+> ```yaml
+> ---
+> type: <Type name>                  # REQUIRED
+> title: <Optional display name>
+> description: <Optional one-line summary>
+> resource: <Optional canonical URI for the underlying asset>
+> tags: [<tag>, <tag>, ...]          # Optional
+> ```
+
+**§4.1, de to bærende definisjonene** (`SPEC.md:194-199`):
+
+> - `description`: A single sentence summarizing the concept. Used by
+>   `index.md` generators, search snippets, and previews.
+> - `tags`: A YAML list of short strings for cross-cutting categorization.
+
+**§11 Conformance** (`SPEC.md:738-742`):
+
+> A bundle is **conformant** with OKF v0.2 if:
+>
+> 1. Every non-reserved `.md` file in the tree contains a parseable YAML
+>    frontmatter block.
+
+### 3.1 Det spec-en ikke sier
+
+`grep -n -i 'nest\|depth\|indent' SPEC.md` gir **null** treff som uttrykker en
+dybde- eller innrykksregel (13 treff, alle i andre betydninger — «flat list of»
+i §9 og §13, «indent» ingen). **SPEC-en har ingen dybderegel.** «Ingen nesting
+forbi dybde 1» er utelukkende vår egen sikkerhetsegenskap. Konformans-gulvet er
+«a parseable YAML frontmatter block» — altså *hele* YAML.
+
+To presiseringer som følger av ordlyden:
+
+- `tags` er definert som «A YAML list» — **ikke** som flow-formen. Skjelettets
+  `[<tag>, ...]` er ett eksempel, ikke formkravet. Blokkformen er like konform.
+- `description` er «A single sentence». En setning brutt over to linjer som
+  foldet plain scalar er samme setning; spec-en stiller ingen linjekrav.
+
+## 4. Måleoppsett
+
+Korpuset er hentet ut ved pinnen, ikke fra arbeidstreet — `_okf-upstream` står
+i dag på `9a15b13`, og `3fcbb9f` er en ekte forgjenger (`git merge-base
+--is-ancestor` → 0), med 23 filer endret i `okf/` mellom dem.
+
+```
+$ git -C _okf-upstream archive 3fcbb9f okf/bundles | tar -x -C scratchpad/corpus
+$ find scratchpad/corpus -name '*.md' ! -name index.md ! -name log.md | wc -l
+53
+```
+
+**Nevner = 53.** Definisjonen er «hver `.md` under `okf/bundles/` som ikke er et
+reservert strukturnavn» — som er nøyaktig §11.1s «every non-reserved `.md` file».
+
+**Positiv kontroll.** Et null-resultat må sjekkes mot et kjent-positivt tilfelle
+før det konsumeres. Følgende dokument parser i dag, mot samme `HEAD`:
+
+```yaml
+type: Metric
+description: Recognized revenue for a period.
+tags:
+  - finance          # INNRYKKET blokksekvens
+  - revenue
+generated: { by: reference_agent/gemini-2.5-pro, at: 2026-06-30T14:00:00Z }
+sources:
+  - id: revenue-policy
+    resource: policies/revenue-recognition.md
+```
+→ `tags` blir `['finance', 'revenue']`, `sources` blir en liste av dicter.
+Instrumentet avviser altså ikke alt; `tags` som blokksekvens **fungerer allerede
+i dag**, forutsatt innrykk.
+
+## 5. Målingen
+
+### 5.1 Formsensus over de 53 (`scratchpad/shapes.py`, exit 0)
+
+| nøkkel | form | antall |
+|---|---|---|
+| `tags` | blokksekvens **uten innrykk** (`- a` i kolonne 0) | **36/53** |
+| `tags` | flow-sekvens `[a, b, c]` | 9/53 |
+| `tags` | enkeltlinje-skalar | 8/53 |
+| `description` | skalar + innrykket fortsettelseslinje | 29/53 |
+| `description` | enkeltlinje-skalar | 24/53 |
+| `generated` | **topp-nivå blokk-mapping** (`  by:` på neste linje) | **44/53** |
+| `generated` | flow-mapping `{ ... }` | 9/53 |
+| `sources` | blokksekvens uten innrykk | 44/53 |
+| `sources` | blokksekvens med innrykk | 5/53 |
+
+### 5.2 Baseline og kandidater (`scratchpad/candidates.py`, exit 0)
+
+Kandidatene er målt ved å **normalisere overflateformen** inn i en form
+parseren allerede godtar, og så importere `parse_frontmatter`. Predikatet er
+aldri re-implementert; bare stavemåten på inputen er skrevet om.
+
+| variant | passerer | dominerende residual |
+|---|---|---|
+| baseline (v1.3.0 som utgitt) | **0/53** | 32× nested mappings, 12× malformed line, 9× flow sequence |
+| **P1** skalar-flow-sekvens | **6/53** | 32× nested mappings, 12× malformed line |
+| **P2** blokksekvens uten innrykk | 0/53 | 44× nested mappings |
+| **P3** foldet plain scalar | 0/53 | 36× malformed line |
+| P1+P3 | 6/53 | 36× malformed line |
+| P1+P2 | 6/53 | 44× nested mappings |
+| **P1+P2+P3** | **6/53** | **44× nested mappings, 3× allowlist** |
+
+De 6 som passerer med P1 er alle i `acme_retail`; de er 6 av de 9 med
+flow-sekvens-`tags`, og de tre siste stoppes av allowlisten
+(`parameters.name` ×2, `not.term` ×1), ikke av `tags`.
+
+### 5.3 Taket
+
+Med alle tre predikatene er taket **6/53**. Residualet er 44× topp-nivå
+blokk-mapping på `generated` og 3× allowlist. En diagnostisk kjøring som også
+normaliserte topp-nivå blokk-mapping til flow-mapping traff neste vegg med én
+gang: 44× «a quoted scalar inside a flow mapping is not a supported form» —
+korpuset skriver `at: '2026-07-10T23:16:06+00:00'` med enkeltfnutter.
+
+## 6. Tre påstander i punkt 44 er målt feil
+
+1. **«The line-flat frontmatter parser has no sequence-value type at all.»**
+   Usant siden 1.3.0. `_consume_block_list` (`okf.py:662`) parser blokklister av
+   rene skalarer, og den positive kontrollen i §4 beviser det. Det som mangler
+   er ikke sekvenstypen, men **innrykkskravet**: `_consume_block_list` krever
+   `raw[:1] in (" ", "\t")`, og korpuset skriver `- a` i kolonne 0.
+   Feilklassen «44/53 rejected as malformed frontmatter line» er i dag 12/53,
+   fordi de øvrige treffer `description`-fortsettelsen først.
+
+2. **«removing `tags` alone lets 4/53 documents pass … removing both … the same
+   4/53.»** Målt i dag: **6/53** med `tags` fjernet, og **6/53** med begge
+   fjernet. Tallet 4 var riktig for 1.2.0-parseren; 1.3.0s `sources`-bærere
+   flyttet to dokumenter til.
+
+3. **«`tags`/`description` reject before `sources` is ever read» → derfor er
+   dette «det ENESTE residualet som blokkerer hele korpuset».** Den slutningen
+   holder ikke. `description` alene løsner **0/53** — fjerner man bare
+   `description`, passerer ingenting. Og lukker man *alle tre* formene, står
+   **44/53** fortsatt på `generated` som topp-nivå blokk-mapping. Den bindende
+   skranken på dette korpuset er altså **ikke** tags/description, men den
+   topp-nivå blokk-mappingen vi bevisst avviser. Punkt 44 overselger sin egen
+   betydning med en faktor på over sju (6 mot 53).
+
+Punkt 44 bør skrives om etter at operatøren har bestemt seg — det er en
+dokumentasjonsendring som hører sammen med predikatvalget, ikke før det.
+
+## 7. Kandidatpredikatene
+
+Ordren spurte etter predikatet som slipper inn en ren skalar-flow-sekvens.
+Det er P1. P2 og P3 tas med fordi målingen viser at P1 alene er en liten
+gevinst, og beslutningen bør se hva naboene koster.
+
+### P1 — flow-sekvens av rene skalarer
+
+Predikat: i `_parse_flow_sequence`, når første ikke-blanke tegn i et element
+ikke er `{`, les elementet som en plain scalar dersom det ikke inneholder noen
+av `{ } [ ] : , " ' #` og ikke er tomt. Blandet sekvens (skalar + mapping)
+avvises, slik blokklisten allerede gjør.
+
+- **(a) Slipper inn:** `tags: [finance, revenue, headline-metric]` — 9/53 i
+  korpuset. Passeringen går fra 0/53 til **6/53**.
+- **(b) Avviser fortsatt:** siterte elementer (`['a', 'b']`), elementer med
+  kolon eller komma i seg, tom sekvens `[]`, uavsluttet `[a, b`, blandet
+  `[a, {b: c}]`, og nestet `[[a]]` — alle på tegn-nivå, uten YAML-semantikk.
+- **(c) Dybde-1:** **bruker den ikke opp.** Elementene er blad; ingen ny
+  nestingsgrad oppstår. Det er samme dybde blokklisten av skalarer allerede har.
+- **(d) Testen som pinner den:** `tags: [a, b]` → `["a", "b"]`; `tags: ['a']`
+  raiser; `tags: [a, {b: c}]` raiser med blandingsfeilen; `tags: []` raiser;
+  `sources: [{ id: x }]` parser uendret (ingen regresjon på G30-bæreren).
+
+### P2 — blokksekvens uten innrykk
+
+Predikat: i `_consume_block_list`, godta også `raw[:1] == "-"` når linjen
+starter med `- ` og forrige toppnøkkel hadde tom verdi.
+
+- **(a) Slipper inn:** `tags:` + `- a` i kolonne 0 — 36/53. Men også `sources:`
+  i samme form — 44/53.
+- **(b) Avviser fortsatt:** alt innholdet i elementet avviser i dag; allowlisten
+  og `_reject_mapping_construct` er uendret.
+- **(c) Dybde-1:** **bruker den ikke opp** for skalarelementer, men den er ikke
+  gratis: den åpner samtidig den ikke-innrykkede blokk-mapping-bæreren for
+  `sources`, hvor elementenes fortsettelseslinjer *er* innrykket. Det er en
+  større flate enn `tags`, og den bør vurderes for seg.
+- **(d) Testen:** `tags:\n- a\n- b` → `["a", "b"]`; `sources:\n- id: x\n  title: y`
+  → én dict; en linje `- a` uten forutgående tom toppnøkkel raiser fortsatt;
+  `tags:\n- a\n- {b: c}` raiser med blandingsfeilen.
+- **Målt effekt alene: 0/53.** Den løsner ingenting uten P1 eller uten at
+  `generated` også åpnes.
+
+### P3 — foldet plain scalar (fortsettelseslinje)
+
+Predikat: etter en toppnøkkel med ikke-tom, ikke-`[`/`{` verdi, slå sammen
+etterfølgende innrykkede linjer som *ikke* starter med `- ` og *ikke* inneholder
+en uquotet `": "`, med ett mellomrom som skjøt.
+
+- **(a) Slipper inn:** `description` brutt over to linjer — 29/53.
+- **(b) Avviser fortsatt:** en innrykket linje som ser ut som `k: v` treffer
+  fremdeles nested-mapping-avvisningen; `- ` treffer fremdeles listeruten.
+  `_reject_dangerous_value` kjører på den sammenslåtte verdien, ikke på
+  fragmentene.
+- **(c) Dybde-1:** **bruker den ikke opp** — resultatet er én skalar.
+  Men den svekker et vern: i dag er *enhver* innrykket linje uten aktiv listenøkkel
+  et avvist nestet uttrykk. Etter P3 er den regelen betinget av at linjen ikke
+  inneholder `": "` — altså samme heuristikk som `_reject_mapping_construct`,
+  gjenbrukt til å *slippe gjennom* i stedet for til å avvise.
+- **(d) Testen:** `description: en setning\n  som fortsetter` → én streng med
+  ett mellomrom; `description: x\n  y: z` raiser fortsatt som nested mapping;
+  `description: x\n  - a` raiser fortsatt.
+- **Målt effekt alene: 0/53.**
+
+## 8. Anbefaling
+
+**P1 alene. Ikke P2, ikke P3, ikke nå.**
+
+Begrunnelsen er tallene, ikke smaken:
+
+- P1 er den eneste av de tre som flytter passeringstallet i det hele tatt
+  (0 → 6). P2 og P3 gir hver for seg **0/53**, og lagt oppå P1 gir de fortsatt
+  **6/53**. De koster parserflate og kjøper null målt konformans.
+- P1 er den minste flaten: den er et tegn-nivå-predikat inne i en funksjon som
+  allerede eksisterer, og den bruker ikke opp dybde-1-regelen.
+- P1 lukker et gap STATE allerede fører som bevisst («`tags: [a, b]`
+  konformansgap»), og den bringer parseren i linje med §4.1s eget skjelett —
+  den ene formen spec-en faktisk skriver ut.
+- P2 og P3 bør ikke besluttes på dette korpuset, fordi korpuset ikke kan skille
+  dem: **44/53 stopper på `generated` som topp-nivå blokk-mapping uansett.**
+  Å bygge P2 og P3 nå ville være å betale for to predikater og måle null.
+
+**Den ærlige konsekvensen, som må sies høyt:** P1 tar korpuset fra 0/53 til
+6/53. Det lukker ikke «hele korpuset». Skal 53/53 nås, er den neste
+beslutningen en helt annen og mye tyngre en — topp-nivå blokk-mapping (44/53)
+pluss siterte skalarer (44/53) — og *den* bruker opp dybde-1-regelen. Det er en
+sikkerhetsbeslutning, ikke en parserdetalj, og den hører ikke i denne ordren.
+
+## 9. Verifiseringslogg
+
+| Påstand | Kommando | Exit | Resultat |
+|---|---|---|---|
+| Nevner = 53 | `find scratchpad/corpus -name '*.md' ! -name index.md ! -name log.md \| wc -l` | 0 | 53 |
+| Pinnen er ekte forgjenger | `git -C _okf-upstream merge-base --is-ancestor 3fcbb9f HEAD` | 0 | ja (HEAD = `9a15b13`) |
+| SPEC-pinnen er ren | `git -C _okf-canonical rev-parse --short HEAD; git status --porcelain` | 0 | `ad30107`, rent tre |
+| SPEC har ingen dybderegel | `grep -n -i 'nest\|depth\|indent' SPEC.md` | 0 | 13 treff, 0 relevante |
+| Baseline 0/53 | `PYTHONPATH=src .venv/bin/python scratchpad/measure.py` | 0 | 0 passerer |
+| Formsensus | `PYTHONPATH=src .venv/bin/python scratchpad/shapes.py` | 0 | tabell §5.1 |
+| Kandidattall | `PYTHONPATH=src .venv/bin/python scratchpad/candidates.py` | 0 | tabell §5.2 |
+| Positiv kontroll | inline, se §4 | 0 | parser, `tags == ['finance','revenue']` |
+| LIMITATIONS-antall | `grep -c '^- \*\*' docs/LIMITATIONS.md` | 0 | 45 |
+| Upushet ved øktstart | `git rev-list --count origin/main..HEAD` | 0 | **0** |
+
+**Ikke målt:** om P1 påvirker ytelse eller ReDoS-marginen — predikatet er
+tegn-for-tegn uten regex, men ingen sveip er kjørt, siden ingen kode er skrevet.
+Kandidatene er målt ved overflatenormalisering, ikke ved en patchet parser: det
+er en trofast simulering av *hva* som slippes inn, men den beviser ikke at en
+implementasjon av P1 avviser nøyaktig (b)-listen. Testene i (d) er det som ville
+pinne det.

docs/ADOPTION-BRIEF.md

@@ -4,7 +4,7 @@
 especially one converging on Google's Open Knowledge Format (OKF v0.1) — and needs
 to decide **when** and **where** to add a write-time ingestion guard.
 
-**Status of the guard:** `v1.1.0`. Stdlib-only core, framework-agnostic. The
+**Status of the guard:** `v1.3.0`. Stdlib-only core, framework-agnostic. The
 exported Python surface is frozen under semver — nothing exported is removed,
 renamed or given a different meaning without a `2.0.0`. Detection behaviour is
 *not* frozen: severities, thresholds and lexicon entries are calibration and move
@@ -142,9 +142,9 @@ live payload:
 python -m llm_ingestion_guard.coverage    # exit 0 = all as documented
 ```
 
-As of `v1.1.0`: **129 / 129 defended classes demonstrated (recall 100%)** and **6 /
+As of `v1.3.0`: **130 / 130 defended classes demonstrated (recall 100%)** and **6 /
 6 documented gaps still hold** (a *closed* gap fails the test, forcing a doc
-update). The matrix is the single source of truth for the test suite (**802
+update). The matrix is the single source of truth for the test suite (**868
 passing**), which also asserts total recall, that every lexicon pattern has a
 case (so the matrix cannot fall behind the lexicon), the full LLM02 secret-egress
 set, and the container-layer front-end (CSV formula-injection, zip-slip/bomb,

docs/BRIEF.md

@@ -3,7 +3,7 @@
 **A reusable, minimal, dependency-light defensive layer for LLM *ingestion*
 pipelines — the write-time siblings of query-time chatbot guardrails.**
 
-Status: implemented — v1.1.0, exported surface frozen under semver. This document defines what the repo contains
+Status: implemented — v1.3.0, exported surface frozen under semver. This document defines what the repo contains
 and why; the stdlib-only core is built and tested (see `README.md` for usage and
 `docs/PLAN.md` for the build order).
 

docs/LIMITATIONS.md

@@ -40,16 +40,31 @@ items; this is the full list, each with the mechanism.
   (an injection in a directory listing is caught) rather than path-rejecting the
   conformant bundle. A front-end materialising individual uploads keeps the opposite
   rule (`allow_reserved=False`): a reserved basename is a listing-shadow and refused.
-- **OKF frontmatter is a restricted grammar: the mapping class has no expressible
-  form.** Gate T2 accepts a line-oriented subset deliberately — full YAML is a larger
-  parse-attack surface than a write-time gate needs. Nested mappings and flow
-  collections (`[a, b]`, `{k: v}`) are *rejected outright*, which fails secure.
-  **All four routes to a mapping fail, each on a different rule** — flow (`{k: v}`)
-  on the disallowed value-start indicator, block (`k:\n  sub: v`) on the
-  nested-mapping check, dotted keys (`k.sub: v`) on the key pattern, and the inline
-  second colon (`k: sub: v`) on the mapping-construct check — so the mapping *class*
-  has no expressible form, rather than one form being preferable to another. What
-  survives is scalars and flat lists of strings. **Two of those routes used to
+- **OKF frontmatter is a restricted grammar: a mapping is expressible through
+  four carriers, all of them key-allowlisted.** Gate T2 accepts a line-oriented
+  subset deliberately — full YAML is a larger parse-attack surface than a
+  write-time gate needs. The admitted carriers are the flow mapping as a value
+  (`generated: { by: x, at: y }`), the flow mapping as a block-list item, the
+  flow *sequence* of flow mappings (`sources: [{ id: a, resource: x }]`) and the
+  block sequence of block mappings (SPEC.md §5.1's own form). **The routes that
+  still fail, each on a different rule:** a top-level block *mapping*
+  (`k:\n  sub: v`) on the nested-mapping check, dotted keys (`k.sub: v`) on the
+  key pattern, the inline second colon (`k: sub: v`) on the mapping-construct
+  check, and a flow sequence of plain *scalars* (`tags: [a, b]`) on the `[`
+  indicator — the sequence carrier is opened for the mapping element and nothing
+  else. **Every carrier is admitted only when every key is on an allowlist**
+  (`by`, `at`, `from`, `to`, `id`, `title`, `author`, `usage_count`,
+  `last_modified` — the keys SPEC.md @ `62432a09` §5.1/§5.2 names inside a mapping,
+  plus `resource` and `usage_window` *under `sources` only*, see below)
+  and every leaf is a plain scalar, itself run through the same value predicates as a
+  top-level scalar. Nested collections, quoted leaves, duplicate keys, an empty or
+  unclosed mapping, and `{a:b}` (which PyYAML 6.0.3 reads as the *key* `a:b`, not as
+  a scalar) all raise. The form is expressible, never trusted: the allowlist
+  inspects every key, which is the property that carried the security when the
+  blanket refusal was doing the enforcing. **A block list may not mix scalar items
+  and mappings** — YAML permits it, but a consumer iterating `sources` and reading
+  `entry.get("id")` gets an `AttributeError` off the first `str`. What else
+  survives is scalars and flat lists of strings. **Two routes used to
   degrade into a string instead of failing, and that defect is closed in `1.1.0`**:
   a block-sequence item carrying exactly one key (`sources:\n  - uri: https://e.com/a`
   yielded the *string* `'uri: https://e.com/a'`) and the inline second colon
@@ -57,7 +72,13 @@ items; this is the full list, each with the mechanism.
   outright). Both parsed "successfully" into the wrong *type*, and a pointer parked
   in one rode through in a key the `resource` allowlist never inspects — mode-b
   `import_bundle` returned WARN and wrote the merged concept verbatim. Both now
-  FAIL_SECURE at T2, before the allowlist is reached. **The boundary is where YAML
+  FAIL_SECURE at T2, before the allowlist is reached. **What closed is the type
+  confusion, not pointer-smuggling as a class:** T3 still inspects `resource` and
+  nothing else, so an honest *string* under another key rides through exactly as
+  before — `attester: attesters/sql_equality.py` is WARN, while the same path
+  under `resource:` FAIL_SECUREs. That is by design (the string is scanned like
+  any other frontmatter value under T1) and it is not what `1.1.0` changed.
+  **The boundary is where YAML
   puts it**, ground-truthed against PyYAML 6.0.3: `": "` and a trailing `":"` open a
   mapping and are refused; a colon carrying neither a space nor a line end
   (`domain:security`, `https://e.com:8443/a`) does not and still parses, as does a
@@ -68,13 +89,76 @@ items; this is the full list, each with the mechanism.
   door A/B persist path, so frontmatter that fails secure on import passes
   `screen_output` unremarked. The grammar therefore bounds what a consumer can *receive*,
   never what a producer can *emit*. Verified identical on 0.2.0 and 0.3.1.
-- **Consequence: an OKF v0.2 concept cannot traverse the external-import path.** Both
-  of v0.2's backward-breaking migration targets are nested — `timestamp` → `generated.at`,
-  and body `# Citations` → a `sources` block list of mappings — so a conformant v0.2
-  concept fails secure at the frontmatter gate. This is the correct direction but it is
-  a compatibility wall, not a policy: v0.2 support requires a deliberate parse-safety
-  decision about widening the grammar, and the dangling-or-substituted `executor`/
-  `attester` pointer question only becomes live once that decision is made.
+- **`sources` passes in both of its spec carriers; the per-entry `usage_window`
+  does not.** The wall used to be total: both of v0.2's backward-breaking migration
+  targets are mappings — `timestamp` → `generated.at`, and body `# Citations` → a
+  `sources` block list of mappings — and a consumer measured **0 of 53** upstream
+  concepts through the gate. `generated.at` stopped being a wall in `1.2.0`;
+  `sources` stopped being one in `1.3.0`, which admits both the block sequence of
+  block mappings (SPEC.md §5.1's own example) and the flow sequence of flow
+  mappings (the form the OKF producers emit, measured 2026-09-02 by two consumers
+  independently). **`resource` is allowlisted inside a `sources` entry and nowhere
+  else.** `1.2.0` left it off on the argument that the parser could not tell
+  `sources[].resource` (§5.1, a citation) from `executor.resource` / `attester.resource`
+  (§10, a pointer to code to be run — door C). That premise was measured false: the
+  owning key is in scope at every call site and was simply never threaded through.
+  It is threaded now, so `executor: [{ resource: skills/run.md }]` and
+  `attester:\n  - resource: …` are refused on the allowlist through *every* carrier,
+  including the two this opened. **What stays refused: a `usage_window` inside a
+  `sources` entry.** SPEC §5.1 permits it per entry ("A single entry MAY carry its
+  own `usage_window`"), and it is a mapping inside a mapping — depth 2, which this
+  parser admits at no key. A bundle using the per-entry override is refused; the
+  shared sibling `usage_window` (the §5.1 example's own form) passes. This is a
+  registered conformance gap, not an oversight: no-nesting-past-depth-1 is a
+  security property, and spending it was not what the fix was for.
+- **`sources[].resource` is scanned as text but never validated as a URL.** T3's
+  https allowlist inspects the *top-level* `resource` and nothing else. It cannot
+  be extended to `sources` entries without over-blocking conformant bundles: SPEC
+  §5.1 explicitly permits a bundle-relative path, a path into `references/`, or a
+  scope descriptor a consumer cannot follow at all (the OKF producers' own golden
+  bundle emits `resource: fixture`). So a `sources` entry may carry
+  `file://`, `javascript:` or any other string; it goes through T1's scan like any
+  other frontmatter value, and nothing else. **A consumer that dereferences
+  `sources[].resource` must validate it itself** — `okf.validate_resource_url` is
+  exported for exactly that. The dangling-or-substituted `executor`/`attester`
+  pointer question stays out of reach separately: both are top-level block
+  mappings, a carrier that is still refused.
+- **`tags` parses in two of the three forms the OKF corpus writes it in;
+  `description`'s folded form still does not — and neither is what caps the
+  corpus.** Re-measured 2026-09-07/08 against the pinned upstream reference
+  bundles (`_okf-upstream/okf` @ `3fcbb9f`, denominator **53** — every
+  non-reserved `.md`, which is SPEC §11.1's own unit) and the pinned SPEC
+  (`_okf-canonical` @ `ad30107`). This entry carried three claims that the
+  measurement showed to be wrong, all three inherited from `1.2.0` and left
+  standing for twelve days after `1.3.0` moved them; they are corrected here.
+  - **A sequence-value type exists** — `_consume_block_list` has parsed block
+    lists of plain scalars since `1.3.0`, and `1.4.0` adds the flow sequence of
+    plain scalars (`tags: [a, b, c]`, SPEC §4.1's own skeleton). What is missing
+    is not the type but the **indentation requirement**: `_consume_block_list`
+    demands a leading space, and the corpus writes `- a` flush at column 0
+    (36/53 of `tags`, 44/53 of `sources`). Shape census of `tags` over the 53:
+    36/53 flush block sequence, 9/53 flow sequence, 8/53 single-line scalar.
+  - **The number is 6/53, not 4/53.** `4` was the `1.2.0` parser's figure.
+    Measured on `1.4.0`, the corpus goes from **0/53 to 6/53** — all six in
+    `acme_retail`, being 6 of the 9 documents whose `tags` is a flow sequence;
+    the other three stop on the mapping-key allowlist (`parameters.name` ×2,
+    `not.term` ×1), not on `tags`.
+  - **`tags`/`description` are not the binding constraint.** `description`
+    alone releases **0/53** — 29/53 write it as a folded plain scalar over an
+    indented second line, which the parser has no continuation-line model for,
+    but removing that obstacle by itself lets nothing through. With all three
+    surface forms closed the corpus still measures **6/53**, because **44/53
+    stop on `generated` written as a top-level block mapping**, and behind that
+    wall sit 44/53 single-quoted scalars inside the mapping. That constraint is
+    refused deliberately: it spends the no-nesting-past-depth-1 rule, which is
+    a security decision and not a parser detail.
+  **Said plainly: `1.4.0` takes this corpus to 6/53, not to 53/53.** The two
+  neighbouring predicates were measured and deliberately **not** built — a
+  flush-left block sequence and a folded plain scalar each release **0/53** on
+  their own, and stacked on top of `1.4.0` they still measure 6/53. They would
+  cost parser surface and buy no measured conformance. The full measurement,
+  including the SPEC quotations and the per-candidate table, is
+  `docs/2026-09-07-limitations-44-maaling.md`.
 - **A persist gate cannot cover execution risk.** OKF v0.2 introduces concepts whose
   purpose is to *name code to be run* (`runtime`, `executor.resource`,
   `attester.resource`). This library answers "is this safe to **store**"; executable
@@ -298,6 +382,21 @@ items; this is the full list, each with the mechanism.
   `test_split_tightens_the_trusted_tier_when_both_carriers_are_present` constructs
   and pins. Read the zero as "not observed in any of the three populations, each
   counted against its own denominator", never as "cannot happen".
+- **The split also LOOSENS the upload door for a lone anchor — the direction it was
+  built for, and the one with a residual worth naming.** Measured as shipped:
+  `<a href="https://ext.example/p">t</a>` on its own emits one
+  `active:raw-html-link` at MEDIUM and disposes `quarantine_review` under
+  `PRESET_USER_UPLOAD` (`warn` under `PRESET_TRUSTED_SOURCE`); through 0.6.1 the
+  same document graded HIGH and `fail_secure`d. An `<img src>` to the same host is
+  untouched — `active:raw-html`, HIGH, `fail_secure`. So an anchor pointing at an
+  attacker-controlled host, arriving on an untrusted upload, is now a human decision
+  rather than a halt. The trade is deliberate and it removes an asymmetry that came
+  from syntax rather than affordance: following an anchor needs a click, exactly like
+  the markdown inline link that has graded MEDIUM since 0.3.1, so the same URL no
+  longer grades two different ways depending on which syntax carries it. It is
+  recorded here so 0.7.0's "frees N documents" is not read as free — what was freed
+  is the click-required class, and MEDIUM is a real grade drop on the door where
+  every finding is trust-escalated.
 - **Raw-HTML findings no longer count end tags, and that moved a published field.**
   Through 0.6.1 `</a>` was active by name on its own, so `count` ran roughly 1.6×
   the opening-tag total (measured on one corpus) and a start/end pair counted 2.
@@ -306,6 +405,44 @@ items; this is the full list, each with the mechanism.
   every document carrying `</a>`, on a field whose meaning did not change. The
   finding count is unaffected: the class still collapses to one finding per class
   per document, and `count` was never a document count.
+- **Which tags the no-URL narrowing may render inert is a judgement about affordance,
+  and no test in this repo can derive it.** `_URL_AFFORDANCE_TAGS` holds the nine
+  names whose entire active affordance *is* the URL they name — `a`, `area`, `img`,
+  `video`, `audio`, `source`, `track`, `frame`, `frameset` — so carrying no URL
+  attribute they name no target and grade inert. Every other name in the active set
+  stays active with no attributes at all, because it does something a URL cannot
+  describe: `<script>` executes its body, `<style>` restyles, `<form>` submits. That
+  boundary is asserted, not measured. A name placed in the set whose affordance does
+  *not* reduce to its URL would go silently invisible, and no corpus can catch it,
+  because what it produces is an absence — the census counts findings, and a tag that
+  stopped firing contributes nothing to count. The fail-secure choice one branch
+  further in holds the other way and is worth reading beside it: a URL attribute whose
+  *value* this module cannot resolve keeps the tag active
+  (`active_content.py:320-324`), a branch the three corpora exercise **0** times.
+  That zero is empirical, so the predicate is written not to depend on it.
+- **"Clean" means *graded, no finding raised* — never *cleaned bytes* — and at one
+  measured consumer's door, `warn` is the floor a document must clear to be
+  persisted rather than rejected.** The word is the library's own: a WARN
+  disposition with nothing to report carries the reason string `"clean: no
+  findings"` (`disposition.py:264`), and this project has repeated that word in the
+  tables it sends consumers. `screen_output` is a judgement API — its
+  `DispositionResult` carries `assessment` / `disposition` / `max_severity` /
+  `reasons`, with no sanitized-text field to read off it. Defanging lives in a
+  separate, deliberate call — `neutralize` — that a caller must invoke itself;
+  nothing upstream of that call transforms a byte. Measured against
+  `llm-ingestion-okf`'s `0.7.0` pin (2026-08-13): `inbox.py:139` sets its persist
+  floor to `warn`, and `inbox.py:323` persists anything carrying that disposition
+  into the bundle; its adapter (`guard_adapter.py:70`) forwards the original
+  extracted text, because nothing upstream ever handed it a transformed one. Four
+  raw-HTML carrier forms the 0.7.0 no-URL narrowing grades inert — an
+  `<a aria-label>` with no `href`, a bare `</a>`, `<Frame>`, `<video />` — verified
+  here (`screen_output(..., PRESET_USER_UPLOAD)`) to dispose `warn, clean: no
+  findings`; at that consumer's door the same four land written into the bundle,
+  carrier present verbatim. Neither library is wrong: `screen_output` never
+  promised transformed bytes, and the consumer never called `neutralize` for them.
+  The gap is in reading "clean" as "sanitized" rather than "no finding raised" — a
+  reading this project's own reports invite, and one that will mislead any caller
+  that persists on `warn` without calling `neutralize` itself.
 - **Measured, document by document: a large minority of *benign* documents do not
   persist unattended at the upload door.** The bullets above bound single rules on
   single URLs. This one bounds the thing a consumer actually feels — how often an
@@ -432,6 +569,23 @@ items; this is the full list, each with the mechanism.
   HIGH under a low-trust preset is `fail_secure`. Report-only means the text is
   never mutated — it does not mean the finding cannot block.
 
+- **The two labels a `<script>` tag raises come from patterns that do not match the
+  same strings.** `hybrid-xss:script-tag` is `<script\b[^><]*>`; `active:raw-html`
+  reads the same tag through `HTML_TAG_RE`, which consumes quoted attribute runs
+  atomically and so tolerates both `<` and `>` inside a quoted value. The lexicon's
+  `<` exclusion is not a modelling choice — it is the 0.3.3 ReDoS fix, and widening
+  it back to `[^>]` restores a quadratic arm (the row below carries the numbers).
+  Measured through both scanners: `<script src="a<b">` raises `active:raw-html` and
+  **no** XSS label, while `<script data-t="a>b">` raises both, the lexicon's match
+  simply ending at the quoted `>`. The disposition never moves — the raw-HTML branch
+  grades `<script>` HIGH with no attributes at all, so every shape here still
+  `fail_secure`s under `PRESET_USER_UPLOAD` — so what the divergence costs is the
+  *label*: a consumer filtering findings on the XSS id sees a subset of the script
+  tags the gate actually caught, and must not read that id as the gate's script-tag
+  census. The residual is practically dead in prose (a `<` inside a script tag's
+  quoted attribute region is not an ordinary shape) and is recorded because the
+  asymmetry is invisible from either scanner alone.
+
 - **A connection-string password longer than 256 chars is not matched.** The
   password run in the `*-connstr` egress patterns is bounded by
   `MAX_CONNSTR_VALUE`; unbounded, it sits in front of a mandatory `@` and makes
@@ -448,24 +602,96 @@ items; this is the full list, each with the mechanism.
   missed; on one preset it is held for review instead of halted.
 
 - **The ReDoS sweep has a measured sensitivity floor, not a clean bill of
-  health.** All 150 compiled patterns across all eleven regex-bearing modules are
+  health.** All 152 compiled patterns across all eleven regex-bearing modules are
   swept arm by arm — payloads synthesised per run from each pattern's own
   skeleton, so `[`, `[system]` and `[system](` are each probed separately rather
   than relying on generic units, and each pattern is timed in the call mode the
   production code uses (`.sub()`/`.finditer()` visit every start position where
   `.match()` cannot). Five patterns were quadratic across 0.3.3 and 0.3.4; all
   are fixed. But the sweep flags on *timing*, and it ignores measurements below a
-  1.5 ms noise floor at N=8000. A quadratic arm sitting just under that floor
-  would still cost **up to ~23 s** at the 1 000 000-char cap. So the claim this
-  sweep supports is "no arm worse than ~23 s at the cap", not "no quadratic arm
-  remains". The method's blind spot is real and has now been demonstrated twice:
-  a generic-payload pass found only one of 0.3.3's two patterns, and 0.3.2's
+  1.5 ms noise floor at N=8000 — process CPU time, re-derived on that instrument
+  (see the clock bullet below) rather than inherited from the wall clock the
+  script used through 1.1.0. A quadratic arm sitting just under that floor would
+  still cost **up to ~23 s** at the 1 000 000-char cap — that figure is arithmetic
+  and not a measurement: a quadratic arm costs the square of the length ratio, and
+  1.5 ms × 125 × 125 is 23.4 s. So the claim this sweep supports is "no arm worse
+  than ~23 s at the cap", not "no quadratic arm remains". The method's blind spot
+  is real and has now been demonstrated twice: a generic-payload pass found only
+  one of 0.3.3's two patterns, and 0.3.2's
   hand-written rows missed all three of 0.3.4's — including one on `sanitize`,
   the first thing every ingested document touches. **Two arm shapes the unit-
   repetition payloads cannot express** are pinned by hand as a result: a tag that
   *closes* around a long body, and a run of plain characters carrying no anchor
   at all.
 
+- **A green ReDoS row is evidence only if it has been seen red, and three rows in
+  this suite had never been.** The class is not a bad bound but a payload that cannot
+  reach the defect, and it leaves the row passing under the vulnerable form too. The
+  sub-agent row is the clearest case: the seed's unbounded lazy run costs per *prefix
+  match*, not per character — each start position where `spawn an agent that ` matches
+  drives its own O(N) scan to end-of-string looking for a capability keyword the
+  payload never supplies, so K prefix matches cost K×O(N), and the bounded
+  `(?:\S+\s+){0,12}?` port caps each scan at 12 tokens for K×O(1). A payload that
+  matches the prefix **once** and then pads pays a single lazy run and is linear
+  however long the pad is — two earlier shapes did exactly that, and the row sat
+  measured-dead at 1.2× until the payload was rebuilt as
+  `"spawn an agent that " * 3000`. (The nesting an older comment blamed is a red
+  herring: the inner `.*?` sits in an optional group, never a repeated one.) Measured
+  through `scan_lexicon` with the seed form patched back in — exponent **1.92** against
+  the shipped **1.01**, and **4.091s vs 0.190s** at 12 000 words, the seed breaking the
+  2.0s bound outright. Two siblings were dead for different reasons.
+  `lexicon-script-tag` had to be given its own N=200 000: at the shared N=100 000 the
+  vulnerable `[^>]` form measured only ~1.2–1.4s — under the 2.0s assert, so the row
+  was green under both forms and proved nothing. And
+  `test_gate_is_bounded_on_the_long_attribute_arm` was killed by **this repo's own
+  narrowing**: 0.7.0 put `<a>` in `_URL_AFFORDANCE_TAGS`, so its `<a ` + 100k + `>`
+  payload became inert and returned *before* the body ever reached the arm the row
+  exists to guard — separation 1.0×, 0.028s and no findings, against 12.475s for the
+  same payload carried by `<script `. The carrier was moved to `<script `, which is
+  active by name with no attributes, so no future URL-shaped narrowing can hollow it
+  out the same way. The general rule the three share: a payload must **deny** the
+  literal the vulnerable run sits in front of — a unit that supplies it matches
+  immediately and never exercises the run. **Nothing but hand measurement finds this
+  class.** The row is green either way, so
+  the suite cannot report its own blind spot, and every bound in it should be read as
+  "verified red under the vulnerable form" only where a comment says it was.
+
+- **Every ReDoS bound in the suite is measured on process CPU time, and so is the
+  sweep that sets the published sensitivity floor.** `tests/redos_clock.py` is the one
+  clock all six test files import — `time.process_time()` — because a blowup is spent
+  cycles while a loaded machine steals wall clock without adding any. On
+  `time.monotonic()` two 0.7.0 rows failed at **2.24s / 3.66s** against a 2.0s bound
+  while two census processes held the CPU, and passed 3/3 on an idle machine: they had
+  been descheduled, not slowed. It lives in one module rather than five copies because
+  `test_output.py::test_the_redos_clock_ignores_time_this_process_did_not_spend` pins
+  one implementation, and four unpinned copies would be free to drift back to a wall
+  clock with nothing going red. **What the CPU clock gives up, stated: a scan that
+  BLOCKS forever burns no CPU, so it would hang the suite instead of failing it.**
+  That is acceptable only because every scanner it measures is pure regex over an
+  in-memory string, with no I/O and no locks — the last wall-clock holdout was retired
+  by auditing its path for anything that could block, not by assumption, and a wall
+  clock guarding a mode that cannot occur still charges the false-red premium
+  (measured there at 21.6s against a 10.0s bound, on a scan that spent 7.6s). **The sweep
+  now runs on the same clock, and closing that divergence bought no sensitivity.**
+  `docs/redos-sweep.py` imports `scan_seconds` instead of timing on
+  `time.monotonic()`, so the 1.5 ms floor at N=8000 and the "~23 s at the cap"
+  figure above are finally in the same currency as the bounds they justify. What
+  the move did *not* do is quiet the sweep, and the floor came back unchanged.
+  Measured over **twelve full runs of all 2585 arms**: the median ratio sits at
+  **1.95–2.03** in every size bucket above 50 µs — the whole surface measures
+  linear — while two-point excursions past the 2.6 flag threshold survive at every
+  magnitude, p99 ratio **2.9–3.3 even above 1 ms**. Flagged arms per run by floor:
+  **6.9 at 0.5 ms, 1.1 at 1.0 ms, 0.33 at 1.5 ms** (0–2 per run), so 1.5 ms is
+  still the knee. Descheduling was never what made this sweep noisy — a ratio
+  computed from two points is. Four distinct arms flagged at the shipped floor
+  across those twelve runs, **each in exactly one of them**, and nine of the twelve
+  runs were clean; eight consecutive runs of the shipped script immediately after a
+  full test run flagged 0–3 arms each, so machine load still moves the count even
+  on a CPU clock. Six flagged arms re-measured over six doublings give exponent
+  **0.97–1.09** and at most 1.2 s at the 1 000 000-char cap. **A single clean run
+  of this sweep is therefore not evidence either** — and neither is a single
+  flagged one.
+
 - **Every surface now bounds its input, but not all of them the same way.**
   `sanitize`, `fence` and `neutralize` raise `OversizeInputError` above
   `MAX_INPUT_CHARS` (1 000 000) rather than returning a partially transformed
@@ -528,6 +754,23 @@ fails the test, forcing this doc to be updated:
   names the outward fetch whatever severity it carries — and must not infer
   "nothing was seen" from a low `Severity`.
 
+- **Under the default action map the assessment axis carries exactly one judgement
+  the disposition does not.** `DEFAULT_ACTION_MAP` sends `NONE` and `LOW` to `WARN`,
+  `ELEVATED` to `QUARANTINE_REVIEW` and `SEVERE` to `FAIL_SECURE` — the last two 1:1.
+  So for any document that carries a finding at all, `assessment` is a relabelling of
+  `disposition` and nothing more; the only thing it adds is *clean* versus *findings
+  present, none dispositive in this context*, which 0.4.0 rendered identically. That
+  collapse is the point (the map is what keeps the separation additive, so a caller
+  ignoring the new axis sees no change), and it is also the limitation: reading
+  `assessment` buys a consumer nothing until it supplies its own `action_map` or needs
+  the clean/low distinction. **The second consequence is on this document.** The
+  published false-positive rates are counts of documents *disposed non-WARN*, and they
+  are a statement about assessed risk only while `NONE` + `LOW` are exactly the WARN
+  pre-image. `tests/test_corpus.py::test_the_published_fp_metric_is_a_risk_statement`
+  pins that equivalence — but it pins it for `DEFAULT_ACTION_MAP`. A caller running its
+  own map makes "disposed non-WARN" a different claim from the one measured here, with
+  nothing in either repo failing to say so.
+
 - **A ZWJ hidden between two emoji is exempt, and ZWNJ's own false-positive
   class is untouched.** U+200D composes emoji (👩‍💻 is WOMAN + ZWJ + PERSONAL
   COMPUTER), so testing it on codepoint membership alone flagged *and stripped*
@@ -546,6 +789,22 @@ fails the test, forcing this doc to be updated:
   not pictographic) and no corpus is available here to verify it against, so it
   is parked as a known false-positive class rather than guessed at.
 
+- **That context test is one predicate on two surfaces, and the symbol carrying it is
+  private.** `sanitize` owns `_is_joiner_in_emoji_sequence`; `output` imports it
+  (`output.py:74`) instead of restating it, because the same defect had to be fixed on
+  both surfaces and a split would let the input side stop flagging while the output
+  side kept hard-blocking — or the reverse, which is how a carrier reaches a persisted
+  artifact after passing the input gate. The agreement is pinned by
+  `tests/test_output.py::test_output_zwj_narrowing_matches_the_sanitize_side`, which
+  asserts `stripped == flagged` across six shapes — half-context on either side, a
+  leading and a trailing joiner, one genuine in-sequence joiner, and a word split.
+  Two things that pin does not give. The six shapes are hand-written rather than
+  drawn from a corpus, so everywhere outside them
+  the surfaces agree by *shared implementation*, not by test — which is the stronger
+  guarantee only for as long as the import survives. And the leading underscore means
+  the predicate is **not** part of the surface frozen under semver: a consumer that
+  imports it is pinning a private name 1.x makes no promise about.
+
 ## Out-of-scope (documented boundary)
 
 Embedding/vector-layer defenses (OWASP LLM08, downstream of persist); multimodal

docs/redos-sweep.py

@@ -37,17 +37,30 @@ import importlib
 import json
 import re
 import sys
-import time
 from dataclasses import dataclass
 from pathlib import Path
 
-SRC = Path(__file__).resolve().parent.parent / "src" / "llm_ingestion_guard"
+ROOT = Path(__file__).resolve().parent.parent
+SRC = ROOT / "src" / "llm_ingestion_guard"
 sys.path.insert(0, str(SRC.parent))
+sys.path.insert(0, str(ROOT / "tests"))
 
 from llm_ingestion_guard.lexicon import load_lexicon  # noqa: E402
+from redos_clock import scan_seconds  # noqa: E402
 
 N1, N2 = 4_000, 8_000
 RATIO_FLAG = 2.6
+# RE-DERIVED on the CPU clock, not inherited from the wall clock this script used
+# through 1.1.0. The clock move fixed false REDS in the suite's bounds; it bought
+# this sweep no sensitivity. Over twelve full runs (2585 arms each) the median
+# ratio is 1.95-2.03 in every size bucket above 50 us -- the whole surface
+# measures linear -- yet two-point excursions past RATIO_FLAG survive at every
+# magnitude (p99 ratio 2.9-3.3 even above 1 ms). Flagged arms per run by floor:
+# 6.9 at 0.5 ms, 1.1 at 1.0 ms, 0.33 at 1.5 ms. The knee is here. Four arms
+# flagged across those twelve runs, each in exactly ONE of them, and six arms
+# that have ever flagged re-measure at exponent 0.97-1.09 over six doublings.
+# Descheduling was never what made this sweep noisy -- a two-point ratio is, so
+# read a clean run and a flagged run with the same suspicion.
 NOISE_FLOOR = 0.0015
 HARD_CAP = 20.0
 
@@ -134,21 +147,35 @@ def build(unit: str, n: int) -> str:
 
 
 def t(rx: re.Pattern[str], text: str, mode: str = "search") -> float:
-    """Time one scan of ``text`` in the mode the production code actually uses."""
+    """Time one scan of ``text`` in the mode the production code actually uses.
+
+    On the SAME clock every ReDoS bound in the suite is measured against --
+    ``tests/redos_clock.py``, process CPU time -- imported rather than restated
+    here, for the reason that module gives: a blowup is spent cycles, and a
+    loaded machine steals wall clock without adding any. Until 1.1.0 this timed
+    on ``time.monotonic()``, which made the floor below and the cap figure
+    derived from it numbers from a different instrument than the bounds they
+    justify. The closure is built BEFORE the clock starts, so only the scan is
+    charged.
+    """
     mode = mode.rstrip("*")
-    start = time.monotonic()
     if mode == "finditer":
-        for _ in rx.finditer(text):
-            pass
+        def scan(s: str) -> None:
+            for _ in rx.finditer(s):
+                pass
     elif mode == "sub":
-        rx.sub("", text)
+        def scan(s: str) -> None:
+            rx.sub("", s)
     elif mode == "match":
-        rx.match(text)
+        def scan(s: str) -> None:
+            rx.match(s)
     elif mode == "fullmatch":
-        rx.fullmatch(text)
+        def scan(s: str) -> None:
+            rx.fullmatch(s)
     else:
-        rx.search(text)
-    return time.monotonic() - start
+        def scan(s: str) -> None:
+            rx.search(s)
+    return scan_seconds(scan, text)
 
 
 # --- targets ----------------------------------------------------------------

llms.txt

@@ -0,0 +1,7 @@
+# llm-ingestion-guard
+
+> Write-time defensive layer for Python pipelines that persist LLM output: sanitize, fence, tool-less quarantined transform, capability isolation, scan before persist, fail-secure.
+
+```bash
+pip install "llm-ingestion-guard @ git+https://git.fromaitochitta.com/open/llm-ingestion-pipeline-security.git@v1.1.0"
+```

pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "hatchling.build"
 
 [project]
 name = "llm-ingestion-guard"
-version = "1.1.0"
+version = "1.4.0"
 description = "Write-time defensive layer for Python pipelines that persist LLM output: sanitize, fence, tool-less quarantined transform, capability isolation, scan before persist, fail-secure."
 readme = "README.md"
 requires-python = ">=3.10"

scratchpad/candidates.py

@@ -0,0 +1,155 @@
+"""What each candidate predicate would admit, measured by normalizing the SURFACE
+form onto a shape the CURRENT parser already accepts. The predicate is imported,
+never re-implemented: only the input's spelling is rewritten."""
+import os, re, sys, collections
+sys.path.insert(0, os.path.join(os.path.dirname(__file__), "..", "src"))
+from llm_ingestion_guard import okf
+
+ROOT = os.path.join(os.path.dirname(__file__), "corpus", "okf", "bundles")
+
+def docs():
+    out = []
+    for dirpath, _d, fns in os.walk(ROOT):
+        for fn in sorted(fns):
+            if fn.endswith(".md") and fn not in ("index.md", "log.md"):
+                out.append(os.path.join(dirpath, fn))
+    return sorted(out)
+
+def split_fm(text):
+    lines = text.split("\n")
+    if not lines or lines[0].strip() != "---":
+        return None, None
+    for i in range(1, len(lines)):
+        if lines[i].strip() == "---":
+            return lines[1:i], lines[i + 1:]
+    return None, None
+
+TOPKEY = re.compile(r"^([A-Za-z0-9_]+):(.*)$")
+
+def n_flow_scalar_seq(fm):
+    """P1: `k: [a, b]` where every element is a plain scalar -> indented block list."""
+    out = []
+    for raw in fm:
+        m = TOPKEY.match(raw)
+        if m and m.group(2).strip().startswith("[") and m.group(2).strip().endswith("]"):
+            inner = m.group(2).strip()[1:-1]
+            elems = [e.strip() for e in inner.split(",")]
+            if inner and all(e and "{" not in e and "}" not in e and "[" not in e for e in elems):
+                out.append("%s:" % m.group(1))
+                out.extend("  - %s" % e for e in elems)
+                continue
+        out.append(raw)
+    return out
+
+def n_flush_block_seq(fm):
+    """P2: a flush-left `- item` run under a bare `k:` -> the same run, indented."""
+    out, i, n = [], 0, len(fm)
+    while i < n:
+        raw = fm[i]
+        m = TOPKEY.match(raw)
+        if m and m.group(2).strip() == "" and i + 1 < n and fm[i + 1][:1] == "-":
+            out.append(raw)
+            i += 1
+            while i < n and (fm[i][:1] == "-" or fm[i][:1] in (" ", "\t")):
+                out.append("  " + fm[i])
+                i += 1
+            continue
+        out.append(raw)
+        i += 1
+    return out
+
+def n_folded_scalar(fm):
+    """P3: `k: text` continued on more-indented plain lines -> one joined line."""
+    out, i, n = [], 0, len(fm)
+    while i < n:
+        raw = fm[i]
+        m = TOPKEY.match(raw)
+        if m and m.group(2).strip() and not m.group(2).strip()[0] in "[{":
+            acc = raw
+            i += 1
+            while i < n and fm[i][:1] in (" ", "\t") and not fm[i].strip().startswith("- "):
+                acc = acc.rstrip() + " " + fm[i].strip()
+                i += 1
+            out.append(acc)
+            continue
+        out.append(raw)
+        i += 1
+    return out
+
+def parses(fm, body):
+    text = "---\n" + "\n".join(fm) + "\n---\n" + "\n".join(body)
+    try:
+        okf.parse_frontmatter(text)
+        return None
+    except okf.OKFFrontmatterError as exc:
+        return str(exc)
+
+COMBOS = [
+    ("baseline (v1.3.0 as shipped)", []),
+    ("P1 scalar flow sequence", [n_flow_scalar_seq]),
+    ("P2 flush block sequence", [n_flush_block_seq]),
+    ("P3 folded plain scalar", [n_folded_scalar]),
+    ("P1+P3", [n_flow_scalar_seq, n_folded_scalar]),
+    ("P2+P3", [n_flush_block_seq, n_folded_scalar]),
+    ("P1+P2", [n_flow_scalar_seq, n_flush_block_seq]),
+    ("P1+P2+P3", [n_flow_scalar_seq, n_flush_block_seq, n_folded_scalar]),
+]
+
+def main():
+    paths = docs()
+    print("denominator: %d documents (pin 3fcbb9f, parser HEAD)" % len(paths))
+    for name, fns in COMBOS:
+        ok = 0
+        residual = collections.Counter()
+        for p in paths:
+            fm, body = split_fm(open(p, encoding="utf-8").read())
+            for fn in fns:
+                fm = fn(fm)
+            err = parses(fm, body)
+            if err is None:
+                ok += 1
+            else:
+                residual[err.split(":")[0]] += 1
+        top = "; ".join("%dx %s" % (c, r) for r, c in residual.most_common(3))
+        print("  %-30s %2d/%d   residual: %s" % (name, ok, len(paths), top or "-"))
+
+main()
+
+# --- diagnostic only (NOT a proposal): size the real binding constraint ---
+def n_block_mapping(fm):
+    """D4: a top-level `k:` followed by indented `a: b` lines -> one flow mapping."""
+    out, i, n = [], 0, len(fm)
+    while i < n:
+        raw = fm[i]
+        m = TOPKEY.match(raw)
+        if m and m.group(2).strip() == "" and i + 1 < n and fm[i + 1][:1] in (" ", "\t") \
+                and not fm[i + 1].strip().startswith("- "):
+            pairs = []
+            i += 1
+            while i < n and fm[i][:1] in (" ", "\t") and not fm[i].strip().startswith("- "):
+                pairs.append(fm[i].strip())
+                i += 1
+            out.append("%s: { %s }" % (m.group(1), ", ".join(pairs)))
+            continue
+        out.append(raw)
+        i += 1
+    return out
+
+print("\n--- diagnostic: what caps the corpus ABOVE the tags/description gap ---")
+for name, fns in [
+    ("D4 top-level block mapping", [n_block_mapping]),
+    ("P1+P2+P3+D4", [n_flow_scalar_seq, n_flush_block_seq, n_folded_scalar, n_block_mapping]),
+]:
+    ok = 0
+    residual = collections.Counter()
+    for p in docs():
+        fm, body = split_fm(open(p, encoding="utf-8").read())
+        for fn in fns:
+            fm = fn(fm)
+        err = parses(fm, body)
+        if err is None:
+            ok += 1
+        else:
+            residual[err.split(":")[0]] += 1
+    top = "; ".join("%dx %s" % (c, r) for r, c in residual.most_common(3))
+    print("  %-30s %2d/53   residual: %s" % (name, ok, top or "-"))

scratchpad/measure.py

@@ -0,0 +1,45 @@
+"""Measure the pinned OKF corpus against the CURRENT parser. No src/ changes."""
+import os, sys, collections
+sys.path.insert(0, os.path.join(os.path.dirname(__file__), "..", "src"))
+from llm_ingestion_guard import okf
+
+ROOT = os.path.join(os.path.dirname(__file__), "corpus", "okf", "bundles")
+
+def docs():
+    out = []
+    for dirpath, _dirnames, filenames in os.walk(ROOT):
+        for fn in sorted(filenames):
+            if not fn.endswith(".md"):
+                continue
+            if fn in ("index.md", "log.md"):
+                continue
+            out.append(os.path.join(dirpath, fn))
+    return sorted(out)
+
+def main():
+    paths = docs()
+    ok, fail = [], []
+    reasons = collections.Counter()
+    for p in paths:
+        with open(p, encoding="utf-8") as fh:
+            text = fh.read()
+        try:
+            okf.parse_frontmatter(text)
+            ok.append(p)
+        except okf.OKFFrontmatterError as exc:
+            fail.append((p, str(exc)))
+            reasons[str(exc).split(":")[0]] += 1
+    print("denominator: %d documents" % len(paths))
+    print("parse OK   : %d" % len(ok))
+    print("parse FAIL : %d" % len(fail))
+    print("--- failure classes ---")
+    for r, c in reasons.most_common():
+        print("%4d  %s" % (c, r))
+    print("--- passing documents (positive control) ---")
+    for p in ok:
+        print("  " + os.path.relpath(p, ROOT))
+    print("--- first 8 failures verbatim ---")
+    for p, e in fail[:8]:
+        print("  %s\n      %s" % (os.path.relpath(p, ROOT), e))
+
+main()

scratchpad/shapes.py

@@ -0,0 +1,59 @@
+"""Shape census of the pinned OKF corpus: which surface form each key is written in."""
+import os, re, sys, collections
+sys.path.insert(0, os.path.join(os.path.dirname(__file__), "..", "src"))
+from llm_ingestion_guard import okf
+
+ROOT = os.path.join(os.path.dirname(__file__), "corpus", "okf", "bundles")
+
+def docs():
+    out = []
+    for dirpath, _d, fns in os.walk(ROOT):
+        for fn in sorted(fns):
+            if fn.endswith(".md") and fn not in ("index.md", "log.md"):
+                out.append(os.path.join(dirpath, fn))
+    return sorted(out)
+
+def split_fm(text):
+    lines = text.split("\n")
+    if not lines or lines[0].strip() != "---":
+        return None, None
+    for i in range(1, len(lines)):
+        if lines[i].strip() == "---":
+            return lines[1:i], lines[i + 1:]
+    return None, None
+
+def key_shape(fm, key):
+    for i, raw in enumerate(fm):
+        m = re.match(r"^(%s):(.*)$" % re.escape(key), raw)
+        if not m:
+            continue
+        val = m.group(2).strip()
+        follow = fm[i + 1] if i + 1 < len(fm) else ""
+        if val.startswith("["):
+            return "flow-sequence"
+        if val.startswith("{"):
+            return "flow-mapping"
+        if val:
+            if follow[:1] in (" ", "\t") and not follow.strip().startswith("- "):
+                return "scalar+continuation"
+            return "single-line-scalar"
+        if follow.strip().startswith("- "):
+            return "block-seq-indented" if follow[:1] in (" ", "\t") else "block-seq-flush"
+        if follow[:1] in (" ", "\t"):
+            return "block-mapping"
+        return "empty"
+    return "ABSENT"
+
+def main():
+    paths = docs()
+    print("denominator: %d" % len(paths))
+    for key in ("tags", "description", "generated", "verified", "sources"):
+        c = collections.Counter()
+        for p in paths:
+            fm, _ = split_fm(open(p, encoding="utf-8").read())
+            c[key_shape(fm, key)] += 1
+        print("\n%s:" % key)
+        for shape, n in c.most_common():
+            print("  %-22s %d/%d" % (shape, n, len(paths)))
+
+main()

src/llm_ingestion_guard/__init__.py

@@ -63,7 +63,7 @@ from .grounding import (
 )
 from . import okf
 
-__version__ = "1.1.0"
+__version__ = "1.4.0"
 
 
 # --- §6 bookends: the two library-side halves around the transform ---------

src/llm_ingestion_guard/coverage.py

@@ -542,8 +542,17 @@ def _build_cases() -> list[Case]:
                     lambda: okf.parse_frontmatter("---\nkey:\n  nested: x\n---\nbody\n"), owasp="LLM10"),
         _raise_case("okf", "T2 frontmatter block scalar", "OKFFrontmatterError",
                     lambda: okf.parse_frontmatter("---\ndesc: |\n  block\n---\nbody\n"), owasp="LLM10"),
-        _raise_case("okf", "T2 frontmatter flow collection", "OKFFrontmatterError",
-                    lambda: okf.parse_frontmatter("---\ntags: [a, b]\n---\nbody\n"), owasp="LLM10"),
+        # The plain-scalar flow sequence is ADMITTED as of 1.4.0 (P1, SPEC §4.1's
+        # own skeleton for `tags`); the class this row measures is the element
+        # shape the parser would have to interpret rather than read -- here a
+        # quoted one, which it would have to strip a quote from to return.
+        _raise_case("okf", "T2 frontmatter flow sequence, quoted element",
+                    "OKFFrontmatterError",
+                    lambda: okf.parse_frontmatter("---\ntags: ['a', 'b']\n---\nbody\n"),
+                    owasp="LLM10"),
+        _raise_case("okf", "T2 mapping key off the allowlist", "OKFFrontmatterError",
+                    lambda: okf.parse_frontmatter(
+                        "---\ngenerated: { by: a, tool: shell }\n---\nbody\n"), owasp="LLM10"),
         _raise_case("okf", "T3 resource non-https (http)", "OKFResourceError",
                     lambda: okf.validate_resource_url("http://insecure.test/x"), owasp="LLM05"),
         _raise_case("okf", "T3 resource data: scheme", "OKFResourceError",

src/llm_ingestion_guard/okf.py

@@ -7,7 +7,8 @@ and feeds scannable text regions into the existing ``sanitize`` / ``scan_output`
 
 T2 — frontmatter parse-safety gate. ``parse_frontmatter`` is a *strict,
 reject-by-default* loader for the minimal OKF frontmatter subset: flat
-``key: value`` scalars plus block ``- item`` lists. Every construct the
+``key: value`` scalars, block ``- item`` lists, and one typed, allowlisted
+mapping form (``{ by: x, at: y }`` — see :func:`_parse_flow_mapping`). Every construct the
 "block anchor/alias DoS + dangerous type coercion" requirement names is refused
 *by construction* — you cannot suffer a billion-laughs alias expansion or a
 ``!!python/object`` coercion if anchors, aliases and explicit tags are rejected
@@ -16,7 +17,9 @@ philosophy, the frontmatter analogue of the ``resource`` reject-gate (T3).
 
 Deliberately NOT a general YAML parser. A security tool whose thesis is
 minimal-dependency should not pull in a full YAML engine whose own features
-(anchors, tags, merges) are the attack surface being defended against. Quoted
+(anchors, tags, merges) are the attack surface being defended against. The one
+mapping form it does admit is admitted key-by-key against an allowlist, not
+parsed generally: the mapping class is expressible, never trusted. Quoted
 scalars are kept verbatim (quotes included) rather than unquoted — the value is
 still scanned as text downstream, so an injection inside a quoted value is not
 lost; richer scalar forms are a future refinement, not a silent parse.
@@ -62,13 +65,57 @@ _KEY_RE = re.compile(r"^[A-Za-z0-9_][A-Za-z0-9_-]*$")
 # A plain OKF scalar cannot *begin* with a YAML structural indicator. Any value
 # starting with one signals an anchor (&), alias (*), explicit tag (!), block
 # scalar (|, >), flow collection ([ ] { }), directive (%) or reserved char
-# (@ `) — all outside the supported subset and all rejected.
+# (@ `) — all outside the supported subset and all rejected. `{` and `[` are
+# tried as the allowlisted mapping form (G3) and the flow sequence of them (G30)
+# FIRST; they reach this predicate only as a leaf inside one, where a nested
+# collection is refused before it can be read.
 _DANGEROUS_VALUE_STARTS = frozenset("&*!|>[]{}%@`")
 # A quoted scalar is a scalar in YAML however many colons it carries, so the
 # mapping check steps aside for one. The quotes are retained rather than
 # stripped — a pre-existing divergence, pinned in tests/test_okf.py.
 _QUOTE_STARTS = frozenset("\"'")
 
+# P1 - what disqualifies a flow-sequence element from being a plain scalar
+# (operator decision, 2026-09-08). Each character is one this parser would have
+# to interpret rather than read: the two quotes (retained, never stripped), the
+# two splitters, the two collection openers and their closers, and the comment
+# indicator. Refusing them is what lets the element be split on commas at the
+# character level without a YAML quote state machine.
+_FLOW_SCALAR_REFUSED = "{}[]:,\"'#"
+
+# G3 - the one mapping form T2 can express (operator decision, 2026-08-21).
+# Every key inside a mapping must be on this allowlist: the form is safe because
+# the allowlist inspects each key, not because mappings became trusted. The keys
+# are the ones OKF v0.2 names inside a mapping - `by`/`at` (SPEC.md @ 62432a09
+# §5.2 `generated`/`verified`) and `from`/`to` (§5.1 `usage_window`), plus the
+# §5.1 `sources`-entry labels.
+_MAPPING_KEY_ALLOWLIST = frozenset({
+    "by", "at", "from", "to", "id", "title", "author", "usage_count",
+    "last_modified",
+})
+
+# G30 - the two §5.1 keys admitted inside a `sources` entry and NOWHERE else
+# (operator decision, 2026-09-02). `resource` is REQUIRED within a `sources`
+# entry, so leaving it off left the whole provenance family unwritable; but the
+# same field name in §10 (`executor.resource`, `attester.resource`) names run
+# instructions and code - the door-C route closed in 1.1.0. 1.2.0 argued the
+# parser could not tell the two apart without parent-key context it did not
+# have. That premise was false: the owning key is in scope at every call site
+# below, it was simply never threaded through. It is threaded now, so the
+# discrimination is structural rather than a judgement about the value.
+# `usage_window` is allowlisted here for accuracy of refusal - §5.1 permits it
+# per entry, and it is then refused on the depth rule (a mapping inside a
+# mapping, which this parser admits at no key) rather than refused as if the
+# key were unknown.
+_SOURCES_ENTRY_KEYS = frozenset({"resource", "usage_window"})
+
+
+def _allowed_mapping_keys(parent_key):
+    """The mapping-key allowlist for a mapping owned by ``parent_key``."""
+    if parent_key == "sources":
+        return _MAPPING_KEY_ALLOWLIST | _SOURCES_ENTRY_KEYS
+    return _MAPPING_KEY_ALLOWLIST
+
 
 class OKFError(Exception):
     """Base class for OKF adapter rejections."""
@@ -106,7 +153,10 @@ def parse_frontmatter(document):
 
     Raises ``OKFFrontmatterError`` on an unterminated fence or any construct
     outside the minimal flat subset (anchors, aliases, explicit tags, merge
-    keys, block scalars, flow collections, nested mappings).
+    keys, block scalars, flow sequences, nested mappings). The single exception
+    is the typed, allowlisted flow mapping (:func:`_parse_flow_mapping`), which
+    parses into a ``dict`` of allowlisted keys with plain-scalar leaves — every
+    other route to a mapping still raises.
     """
     lines = document.split("\n")
     if not lines or lines[0].strip() != _FENCE:
@@ -149,13 +199,31 @@ def _scannable_regions(frontmatter, body):
     """The text regions of a concept that carry attacker-controlled content."""
     regions = [body]
     for value in frontmatter.values():
-        if isinstance(value, list):
-            regions.extend(value)
-        elif value:
-            regions.append(value)
+        regions.extend(_value_regions(value))
     return regions
 
 
+def _value_regions(value):
+    """Every scannable leaf of one frontmatter value.
+
+    A mapping value (G3) is a new *shape* on this surface, not a new exemption:
+    its leaves are scanned exactly like a scalar or a list item, so an injection
+    parked in ``generated: { by: ... }`` reaches ``scan_output`` like any other
+    frontmatter text. The same holds for a *list* of mappings (G30, ``sources``),
+    which this function already flattens through its list branch. Mapping *keys*
+    are not scanned because they cannot carry attacker text - the allowlist
+    admits a fixed, per-parent name set and nothing else.
+    """
+    if isinstance(value, dict):
+        return [leaf for leaf in value.values() if leaf]
+    if isinstance(value, list):
+        regions = []
+        for item in value:
+            regions.extend(_value_regions(item))
+        return regions
+    return [value] if value else []
+
+
 def validate_concept_path(path, *, allow_reserved=False):
     """Validate a bundle-relative concept path and return its concept-ID.
 
@@ -575,10 +643,22 @@ def _parse_flat(fm_lines):
             raise OKFFrontmatterError("invalid frontmatter key: %r" % key)
 
         if value == "":
-            items, i = _consume_block_list(fm_lines, i + 1)
+            items, i = _consume_block_list(fm_lines, i + 1, key)
             result[key] = items if items is not None else ""
             continue
 
+        mapping = _parse_flow_mapping(value, key)
+        if mapping is not None:
+            result[key] = mapping
+            i += 1
+            continue
+
+        sequence = _parse_flow_sequence(value, key)
+        if sequence is not None:
+            result[key] = sequence
+            i += 1
+            continue
+
         _reject_dangerous_value(value)
         _reject_mapping_construct(value)
         result[key] = value
@@ -587,14 +667,26 @@ def _parse_flat(fm_lines):
     return result
 
 
-def _consume_block_list(fm_lines, start):
+def _consume_block_list(fm_lines, start, parent_key=None):
     """Consume ``  - item`` lines following a bare ``key:``.
 
-    Returns ``(items, next_index)`` — ``items`` is ``None`` (and ``next_index``
+    Returns ``(items, next_index)`` - ``items`` is ``None`` (and ``next_index``
     unchanged) when no list item follows, so the caller can treat the key as an
     empty scalar and let the next line trip the nested-structure guard.
+
+    An item is one of three shapes, decided by the item text alone: a flow
+    mapping (G3), a block mapping (G30 - an unquoted ``key: value`` opening a
+    run of more-indented sibling entries, which is SPEC.md §5.1's own carrier
+    for ``sources``), or a plain scalar. ``parent_key`` is the key that owns the
+    list; it decides the mapping-key allowlist, which is how ``sources[].resource``
+    is admitted while ``executor``/``attester`` ``resource`` stays refused.
+
+    A list may not mix scalars and mappings. YAML permits it, but a consumer
+    iterating ``sources`` and reading ``entry.get("id")`` gets an
+    ``AttributeError`` off the first ``str`` - refusing is the cheaper failure.
     """
     items = []
+    kinds = set()
     i = start
     n = len(fm_lines)
     while i < n:
@@ -603,19 +695,113 @@ def _consume_block_list(fm_lines, start):
         if stripped == "" or stripped.startswith("#"):
             i += 1
             continue
-        if raw[:1] in (" ", "\t") and stripped.startswith("- "):
-            item = stripped[2:].strip()
-            _reject_dangerous_value(item)
-            _reject_mapping_construct(item)
-            items.append(item)
+        if not (raw[:1] in (" ", "\t") and stripped.startswith("- ")):
+            break
+
+        item = stripped[2:].strip()
+
+        mapping = _parse_flow_mapping(item, parent_key)
+        if mapping is not None:
+            items.append(mapping)
+            kinds.add("mapping")
             i += 1
             continue
-        break
+
+        entry = _block_mapping_entry(item)
+        if entry is not None:
+            mapping, i = _consume_block_mapping(fm_lines, i + 1, entry, parent_key)
+            items.append(mapping)
+            kinds.add("mapping")
+            continue
+
+        _reject_dangerous_value(item)
+        _reject_mapping_construct(item)
+        items.append(item)
+        kinds.add("scalar")
+        i += 1
+
+    if len(kinds) > 1:
+        raise OKFFrontmatterError(
+            "a block list may not mix scalar items and mappings: %r" % (parent_key,)
+        )
     if not items:
         return None, start
     return items, i
 
 
+def _block_mapping_entry(text):
+    """Read ``text`` as one ``key: value`` block-mapping entry, or return ``None``.
+
+    The trigger is deliberately the same shape ``_reject_mapping_construct``
+    uses to *refuse* a scalar: an unquoted ``": "``. What changes in 1.3.0 is
+    only what happens next - the entry is admitted key-by-key against the
+    allowlist instead of refused wholesale. Every shape that is a scalar to
+    PyYAML stays one here: a quoted item, a colon with no space
+    (``domain:security``, ``https://e.com:8443/a``) and a trailing colon all
+    return ``None`` and fall through to the unchanged scalar rules.
+    """
+    if not text or text[0] in _QUOTE_STARTS:
+        return None
+    key, sep, leaf = text.partition(": ")
+    if not sep:
+        return None
+    key = key.strip()
+    if not _KEY_RE.match(key):
+        return None
+    return key, leaf.strip()
+
+
+def _consume_block_mapping(fm_lines, start, first_entry, parent_key):
+    """Consume the sibling entries of a block mapping opened by a ``- `` item.
+
+    Returns ``(mapping, next_index)``. A sibling is an indented line that does
+    not open a new list item; the run ends at a blank line, a comment, a new
+    ``- `` item, or a line at column zero. Depth is capped at one by giving the
+    leaves the *unchanged* scalar predicates: a nested collection opens with
+    ``{`` or ``[`` and is refused by ``_reject_dangerous_value``, and a further
+    block level is refused by ``_reject_mapping_construct``.
+    """
+    allowed = _allowed_mapping_keys(parent_key)
+    mapping = {}
+    _admit_mapping_entry(mapping, first_entry[0], first_entry[1], allowed, parent_key)
+
+    i = start
+    n = len(fm_lines)
+    while i < n:
+        raw = fm_lines[i]
+        stripped = raw.strip()
+        if stripped == "" or stripped.startswith("#"):
+            break
+        if raw[:1] not in (" ", "\t") or stripped.startswith("- "):
+            break
+        entry = _block_mapping_entry(stripped)
+        if entry is None:
+            _reject_dangerous_value(stripped)
+            _reject_mapping_construct(stripped)
+            raise OKFFrontmatterError(
+                "a block-mapping entry must be 'key: value': %r" % (raw,)
+            )
+        _admit_mapping_entry(mapping, entry[0], entry[1], allowed, parent_key)
+        i += 1
+    return mapping, i
+
+
+def _admit_mapping_entry(mapping, key, leaf, allowed, parent_key):
+    """Admit one mapping entry, or raise. The single gate both carriers pass."""
+    if not _KEY_RE.match(key):
+        raise OKFFrontmatterError("invalid mapping key: %r" % (key,))
+    if key not in allowed:
+        raise OKFFrontmatterError(
+            "mapping key %r is not on the OKF mapping allowlist under %r"
+            % (key, parent_key)
+        )
+    if key in mapping:
+        raise OKFFrontmatterError("duplicate mapping key %r" % (key,))
+    _reject_dangerous_value(leaf)
+    _reject_mapping_construct(leaf)
+    mapping[key] = leaf
+
+
 def _reject_dangerous_value(value):
     if value and value[0] in _DANGEROUS_VALUE_STARTS:
         raise OKFFrontmatterError(
@@ -627,8 +813,10 @@ def _reject_dangerous_value(value):
 def _reject_mapping_construct(value):
     """Reject a scalar that YAML reads as a mapping rather than as a string.
 
-    T2 gives the mapping *class* no expressible form — flow, nested-block and
-    dotted-key routes all raise. Two routes used to escape that by degrading
+    T2 gives the mapping *class* exactly one expressible form, the typed
+    allowlisted flow mapping (G3); the nested-block and dotted-key routes still
+    raise, and this predicate is what keeps them raising — both at the top level
+    and on a leaf *inside* an admitted mapping. Two routes used to escape by degrading
     into a string instead: a block-sequence item carrying exactly one key
     (``- uri: x``), and an inline second colon (``attester: resource: x``).
     Both parsed "successfully" into the wrong *type*, and a pointer parked in
@@ -648,3 +836,202 @@ def _reject_mapping_construct(value):
         raise OKFFrontmatterError(
             "a mapping is not expressible in OKF frontmatter: %r" % (value,)
         )
+
+
+def _parse_flow_mapping(value, parent_key=None):
+    """Parse ``{ key: value, ... }`` into a typed dict, or refuse it (G3).
+
+    Returns ``None`` when ``value`` does not open a flow mapping, so the caller
+    falls through to the unchanged scalar rules. Otherwise the value either
+    parses into a ``dict`` of allowlisted keys with plain-scalar leaves, or
+    raises - it never degrades into a string, which is the defect closed in
+    1.1.0 and not reopened here.
+
+    Why the mapping class needed *a* form at all: OKF v0.2 writes its whole
+    trust and provenance layer as mappings, and SPEC.md @ ``62432a09`` uses flow
+    form in its own examples (§5.1 ``usage_window``, §5.2 ``generated`` /
+    ``verified``). §11 goes further than "should": a consumer *MUST* treat a
+    bare ``verified`` mapping as a one-element list - a rule that presupposes
+    the mapping parses. With no form, 0 of 53 upstream concepts reached the
+    gate, and no threshold would have changed that.
+
+    Why this form is safe: the allowlist inspects **every key**, which is the
+    property that actually carried the security in T2 - the blanket refusal was
+    the enforcement, not the point. Admitted, ground-truthed against PyYAML
+    6.0.3:
+
+    - one flow mapping per value, closed on the same line (``{ a: b }``);
+    - keys on :data:`_MAPPING_KEY_ALLOWLIST` and matching ``_KEY_RE``, no
+      duplicates - PyYAML resolves a duplicate last-wins, which is a way to
+      show one claim and mean another;
+    - plain-scalar leaves only, each run through the *unchanged*
+      ``_reject_dangerous_value`` / ``_reject_mapping_construct`` predicates, so
+      a leaf can no more open an anchor, a tag or a nested mapping than a
+      top-level scalar can.
+
+    Refused, each on its own rule: nested collections (``{ a: { b: c } }``,
+    ``{ a: [1] }``), quoted leaves, an empty mapping, an unclosed or
+    trailing-junk value (``{ a: b } x``, which PyYAML also refuses), a key
+    outside the allowlist, and ``{a:b}`` - which PyYAML reads as the *key*
+    ``a:b``, not as a scalar, and which the required ``": "`` separator catches.
+
+    Two deliberate divergences from PyYAML, both toward refusal: a quoted leaf
+    (``{ title: 'a, b' }``) and a trailing comment (``{ a: b } # note``) are
+    conformant YAML that this rejects. Splitting quoted commas correctly needs a
+    quote state machine whose failure mode is *accepting* something YAML would
+    refuse; refusing is the cheaper side to be wrong on, and the keys that
+    plausibly need a comma (``title``, ``author``) only occur inside ``sources``
+    entries, whose block-sequence carrier is refused anyway.
+    """
+    if not value or value[0] != "{":
+        return None
+    if not value.endswith("}"):
+        raise OKFFrontmatterError(
+            "a flow mapping must be closed by '}' on the same line: %r" % (value,)
+        )
+
+    inner = value[1:-1].strip()
+    if inner.endswith(","):  # a trailing comma is legal YAML; one, and only one
+        inner = inner[:-1].strip()
+    if not inner:
+        raise OKFFrontmatterError("an empty flow mapping carries nothing: %r" % (value,))
+    for char in "{}[]":
+        if char in inner:
+            raise OKFFrontmatterError(
+                "a flow mapping admits scalar leaves only, not %r: %r" % (char, value)
+            )
+    for quote in _QUOTE_STARTS:
+        if quote in inner:
+            raise OKFFrontmatterError(
+                "a quoted scalar inside a flow mapping is not a supported form: %r"
+                % (value,)
+            )
+
+    allowed = _allowed_mapping_keys(parent_key)
+    mapping = {}
+    for entry in inner.split(","):
+        entry = entry.strip()
+        key, sep, leaf = entry.partition(": ")
+        if not sep:
+            raise OKFFrontmatterError(
+                "a flow-mapping entry must be 'key: value': %r" % (entry,)
+            )
+        _admit_mapping_entry(mapping, key.strip(), leaf.strip(), allowed, parent_key)
+    return mapping
+
+
+def _parse_flow_sequence(value, parent_key=None):
+    """Parse ``[{ ... }, { ... }]`` into a list of typed dicts, or refuse it (G30).
+
+    Returns ``None`` when ``value`` does not open a flow sequence, so the caller
+    falls through to the unchanged scalar rules - where ``[`` is still a
+    disallowed indicator. This carrier is opened for the flow-mapping element
+    and nothing else: it is the form the OKF producers emit for ``sources``
+    (measured 02.09 against llm-ingestion-okf's golden bundle, where a
+    one-element sequence raised on the ``[`` just as a two-element one did).
+
+    As of 1.4.0 it also carries a sequence of plain *scalars*
+    (``tags: [a, b, c]``) - SPEC.md §4.1's own skeleton for ``tags``, and the
+    one candidate form measured to move the upstream corpus at all (0/53 ->
+    6/53 against ``_okf-upstream`` @ 3fcbb9f, denominator 53; see
+    docs/2026-09-07-limitations-44-maaling.md). The quoting and comma-splitting
+    problem that kept it refused is answered by refusing the characters that
+    create it rather than by parsing them: an element is a plain scalar only if
+    it is non-empty and carries none of ``{ } [ ] : , " ' #``, and it then
+    passes the unchanged scalar indicator rule. Everything needing YAML
+    semantics to split or unquote correctly still raises.
+
+    A mapping element is split on ``}`` rather than on commas, which is sound
+    precisely because ``_parse_flow_mapping`` admits no nested collection: a
+    ``}`` inside an element cannot occur, so the first ``}`` after ``{`` always
+    closes it. A scalar element runs to the next comma, which cannot occur
+    inside one. Anything between elements that is not a separating comma is
+    refused, which is what makes trailing junk fail rather than parse.
+
+    A sequence may not mix the two, for the reason the block list may not: a
+    consumer iterating the value and reading ``entry.get("id")`` gets an
+    ``AttributeError`` off the first ``str``. The mixing verdict is reached
+    before the element is parsed, so the caller is told about the mix rather
+    than about a key the allowlist would have complained of instead.
+    """
+    if not value or value[0] != "[":
+        return None
+    if not value.endswith("]"):
+        raise OKFFrontmatterError(
+            "a flow sequence must be closed by ']' on the same line: %r" % (value,)
+        )
+
+    inner = value[1:-1].strip()
+    if not inner:
+        raise OKFFrontmatterError("an empty flow sequence carries nothing: %r" % (value,))
+
+    items = []
+    kinds = set()
+    i = 0
+    n = len(inner)
+    while True:
+        while i < n and inner[i] in " \t":
+            i += 1
+        if i >= n:
+            break
+        if inner[i] == "{":
+            _refuse_mixed_flow_sequence(kinds, "mapping", value)
+            close = inner.find("}", i)
+            if close == -1:
+                raise OKFFrontmatterError(
+                    "an unclosed flow mapping inside a flow sequence: %r" % (value,)
+                )
+            items.append(_parse_flow_mapping(inner[i:close + 1], parent_key))
+            i = close + 1
+        else:
+            _refuse_mixed_flow_sequence(kinds, "scalar", value)
+            end = i
+            while end < n and inner[end] != ",":
+                end += 1
+            items.append(_flow_sequence_scalar(inner[i:end].strip(), value))
+            i = end
+        while i < n and inner[i] in " \t":
+            i += 1
+        if i >= n:
+            break
+        if inner[i] != ",":
+            raise OKFFrontmatterError(
+                "trailing junk after a flow-sequence element: %r" % (value,)
+            )
+        i += 1
+    return items
+
+
+def _refuse_mixed_flow_sequence(kinds, kind, value):
+    kinds.add(kind)
+    if len(kinds) > 1:
+        raise OKFFrontmatterError(
+            "a flow sequence may not mix scalar items and mappings: %r" % (value,)
+        )
+
+
+def _flow_sequence_scalar(element, value):
+    """Read one flow-sequence element as a plain scalar, or refuse it (P1).
+
+    Character-level, with no YAML semantics: the element must be non-empty and
+    carry none of :data:`_FLOW_SCALAR_REFUSED`. That set is not a style rule -
+    each member is a character whose meaning this parser would have to guess at.
+    A quote would have to be stripped (this parser retains quotes, so it would
+    hand back a different value than YAML reads); a comma or a colon would have
+    to be split on; a bracket or a brace would open a second collection level,
+    which no carrier here admits; a ``#`` opens a comment. The unchanged
+    indicator rule then applies to what is left, exactly as it does to a
+    block-list item, so an anchor or an alias is no more a scalar here.
+    """
+    if not element:
+        raise OKFFrontmatterError(
+            "an empty element in a flow sequence carries nothing: %r" % (value,)
+        )
+    for char in _FLOW_SCALAR_REFUSED:
+        if char in element:
+            raise OKFFrontmatterError(
+                "a flow-sequence scalar admits plain scalars only, not %r: %r"
+                % (char, value)
+            )
+    _reject_dangerous_value(element)
+    return element

tests/test_docs_measurement_scripts.py

@@ -28,10 +28,12 @@ from pathlib import Path
 
 import pytest
 
+import redos_clock
 from llm_ingestion_guard import Disposition, PRESET_USER_UPLOAD, Risk, screen_output
 from llm_ingestion_guard import active_content as ac
 
 _DOCS = Path(__file__).resolve().parent.parent / "docs"
+_LIMITATIONS = _DOCS / "LIMITATIONS.md"
 
 
 def _load(filename: str):
@@ -41,6 +43,7 @@ def _load(filename: str):
     spec = importlib.util.spec_from_file_location(name, path)
     assert spec and spec.loader, f"cannot load {path}"
     module = importlib.util.module_from_spec(spec)
+    sys.modules[name] = module  # @dataclass resolves its own module via sys.modules
     spec.loader.exec_module(module)
     return module
 
@@ -49,6 +52,7 @@ def _load(filename: str):
 # script's import list fails the whole file loudly instead of one quiet test.
 fp_sweep = _load("fp-sweep.py")
 census = _load("rawhtml-census.py")
+redos_sweep = _load("redos-sweep.py")
 
 
 # --- docs/fp-sweep.py --------------------------------------------------------
@@ -245,6 +249,38 @@ def test_census_production_row_equals_its_shipped_candidate(name, attrs):
     assert candidate(name, attrs) == ac.active_tag_class(name, attrs)
 
 
+# --- docs/redos-sweep.py ------------------------------------------------------
+
+
+def test_redos_sweep_times_on_the_suite_clock_not_a_reimplementation():
+    # Until 1.1.0 this script timed on `time.monotonic()`, a different instrument
+    # than every ReDoS bound in the suite. `t()` must call the shared
+    # `scan_seconds` — imported, not restated — and the module must not import
+    # `time` itself, else a drift back to a wall clock would go unnoticed here.
+    assert redos_sweep.scan_seconds is redos_clock.scan_seconds
+    assert not hasattr(redos_sweep, "time"), "module must not import time itself"
+
+
+def test_redos_sweep_floor_and_flag_match_the_published_numbers():
+    # docs/LIMITATIONS.md publishes the 1.5 ms floor and the 2.6 flag ratio this
+    # script derives from twelve full runs. Pin both sides: the constants, and
+    # that the doc still states the same numbers — either drifting alone is a bug.
+    assert redos_sweep.NOISE_FLOOR == 0.0015
+    assert redos_sweep.RATIO_FLAG == 2.6
+    text = _LIMITATIONS.read_text(encoding="utf-8")
+    assert "1.5 ms noise floor" in text
+    assert "2.6 flag threshold" in text
+
+
+def test_redos_sweep_collector_covers_152_patterns_across_11_tables():
+    # The count docs/LIMITATIONS.md carries as "all 152 compiled patterns across
+    # all eleven regex-bearing modules". A pattern added or removed in `src/`
+    # without re-measuring would drift the doc's claim silently otherwise.
+    assert len(redos_sweep.TABLES) == 11
+    total = sum(len(collect()) for collect in redos_sweep.TABLES.values())
+    assert total == 152
+
+
 # --- both scripts: the argument-less contract --------------------------------
 
 

tests/test_okf.py

@@ -116,9 +116,13 @@ def test_rejects_unterminated_frontmatter():
 
 
 def test_rejects_flow_collection():
-    # inline flow collections are outside the supported subset -> reject, don't
-    # silently mis-parse the bracket string as a scalar.
-    doc = "---\ntype: table\ntags: [pii, customers]\n---\nbody\n"
+    # An inline flow collection this parser cannot read without guessing ->
+    # reject, don't silently mis-parse the bracket string as a scalar. The
+    # QUOTED elements are what makes this row stay red: the plain-scalar form
+    # (`tags: [pii, customers]`) is admitted as of 1.4.0 -- see the P1 block at
+    # the foot of this file -- while a quoted element would have to be stripped
+    # to be read, and this parser retains quotes.
+    doc = "---\ntype: table\ntags: ['pii', 'customers']\n---\nbody\n"
     with pytest.raises(OKFFrontmatterError):
         parse_frontmatter(doc)
 
@@ -548,9 +552,15 @@ _V02_REJECTED = [
     ("generated (nested)", "generated:\n  at: 2026-07-26T10:00:00Z\n"),
     ("executor (nested)", "executor:\n  resource: skills/run-on-bq.md\n"),
     ("attester (nested)", "attester:\n  resource: attesters/sql_equality.py\n"),
-    ("sources (block list of mappings)",
+    # The carrier is admitted as of 1.3.0 (G30); this row now measures the KEY
+    # SET - `uri`/`kind` are producer-invented, not SPEC §5.1 - and stays red for
+    # that reason. SPEC's own §5.1 keys parse; see the G30 block at the foot.
+    ("sources (block list, off-allowlist keys)",
      "sources:\n  - uri: https://e.com/a\n    kind: doc\n"),
-    ("flow sequence", "tags: [a, b, c]\n"),
+    # The plain-scalar flow sequence is admitted as of 1.4.0 (P1); this row now
+    # measures the QUOTED element, which stays refused for the same reason it
+    # always did -- reading it would mean stripping a quote this parser retains.
+    ("flow sequence, quoted elements", "tags: ['a', 'b']\n"),
     ("flow mapping", "executor: {resource: skills/run.md}\n"),
 ]
 
@@ -569,6 +579,8 @@ _V02_ADMITTED = [
     ("status/stale_after", "status: active\nstale_after: 2026-12-01\n"),
     ("verified bool", "verified: true\n"),
     ("block sequence of scalars", "tags:\n  - alpha\n  - beta\n"),
+    # SPEC §4.1's own skeleton for `tags`, admitted as of 1.4.0 (P1).
+    ("flow sequence of scalars", "tags: [alpha, beta]\n"),
 ]
 
 
@@ -586,6 +598,8 @@ def test_v02_flat_frontmatter_still_parses(cid, fm):
 
 _DEGRADED_TO_STRING = [
     # (id, frontmatter, what PyYAML 6.0.3 makes of it)
+    # Still red after G30 opened the block-mapping carrier: `uri` is not a §5.1
+    # key, so the item is refused by the allowlist instead of by the grammar.
     ("one key per item", "sources:\n  - uri: https://e.com/a\n", "[{'uri': ...}]"),
     ("item, trailing colon", "sources:\n  - uri:\n", "[{'uri': None}]"),
     ("inline double colon", "attester: resource: attesters/sql_equality.py\n", "parse error"),
@@ -596,9 +610,11 @@ _DEGRADED_TO_STRING = [
 @pytest.mark.parametrize("cid,fm,yaml_reads_as", _DEGRADED_TO_STRING,
                          ids=[c[0] for c in _DEGRADED_TO_STRING])
 def test_a_mapping_construct_never_degrades_into_a_string(cid, fm, yaml_reads_as):
-    # The mapping *class* has no expressible form through T2 — so a mapping
-    # construct must RAISE, never parse "successfully" into the wrong type. A
-    # consumer reading frontmatter["sources"][0].get("uri") must not be handed a str.
+    # None of these shapes is the one form T2 admits (G3, the allowlisted flow
+    # mapping) — so each must RAISE, never parse "successfully" into the wrong
+    # type. A consumer reading frontmatter["sources"][0].get("uri") must not be
+    # handed a str, and that holds whether the mapping class has no expressible
+    # form or one.
     with pytest.raises(OKFFrontmatterError):
         parse_frontmatter(f"---\nid: x\n{fm}---\n\nbody\n")
 
@@ -651,13 +667,20 @@ def test_pointer_in_a_degraded_mapping_no_longer_reaches_the_consumer_tree(cid,
     assert result.disposition is Disposition.FAIL_SECURE, "hole reopened — see LIMITATIONS.md"
 
 
-def test_every_route_to_a_mapping_fails_on_a_different_rule():
-    # The v0.2 wall is not a choice between two forms where one is better: ALL three
-    # ways to express a mapping fail, each on its own rule, so the mapping *class* has
-    # no expressible form through T2. v0.2's `generated` IS a mapping (`by` required
-    # when present), so it cannot be expressed at all.
+def test_the_expressible_mapping_routes_are_the_ones_the_allowlist_inspects():
+    # Was `test_exactly_one_route_to_a_mapping_is_expressible` (1.2.0), and before
+    # that ALL FOUR routes failed so the mapping *class* had no expressible form.
+    # There are four expressible carriers as of 1.3.0 - flow mapping as a value,
+    # flow mapping as a list item, flow sequence of flow mappings, block sequence
+    # of block mappings - and the criterion that admits them is unchanged: each
+    # hands the allowlist every key. The routes below stay shut for the same
+    # reason, each on its own rule: a top-level block MAPPING (not a sequence),
+    # a dotted key, and an inline second colon give the allowlist nothing to
+    # inspect.
+    assert parse_frontmatter("---\nid: x\ngenerated: { by: x, at: y }\n---\n\nbody\n")[0][
+        "generated"] == {"by": "x", "at": "y"}
+
     routes = {
-        "flow":   "generated: { by: x, at: y }\n",
         "block":  "generated:\n  by: x\n",
         "dotted": "generated.by: x\n",
         "inline": "generated: by: x\n",
@@ -667,11 +690,10 @@ def test_every_route_to_a_mapping_fails_on_a_different_rule():
         with pytest.raises(OKFFrontmatterError) as exc:
             parse_frontmatter(f"---\nid: x\n{fm}---\n\nbody\n")
         errors[name] = str(exc.value)
-    assert "indicator" in errors["flow"]
     assert "nested mappings" in errors["block"]
     assert "key" in errors["dotted"]
     assert "mapping" in errors["inline"]
-    assert len(set(errors.values())) == 4, "routes must fail distinctly, not collapse"
+    assert len(set(errors.values())) == 3, "routes must fail distinctly, not collapse"
 
 
 _BLOCK_LIST_ITEM_SHAPES = [
@@ -690,17 +712,35 @@ def test_block_lists_admitted_by_item_shape(cid, fm, expected):
     assert parse_frontmatter(f"---\nid: x\n{fm}---\n\nbody\n")[0][key] == expected
 
 
-def test_two_keys_per_item_is_where_the_block_list_hard_rejects():
+def test_the_block_list_rejects_on_the_key_set_not_on_arity():
+    # Was `test_two_keys_per_item_is_where_the_block_list_hard_rejects`: a
+    # two-key item was refused because the block mapping had no expressible form
+    # at all. G30 gives it one (SPEC.md §5.1's own carrier), so arity is no
+    # longer the boundary - the key set is. The same two-key item parses when its
+    # keys are §5.1's, and still raises when one of them is not.
+    parsed = parse_frontmatter(
+        "---\nid: x\nsources:\n  - id: a\n    resource: file://x\n---\n\nbody\n"
+    )[0]
+    assert parsed["sources"] == [{"id": "a", "resource": "file://x"}]
     with pytest.raises(OKFFrontmatterError):
         parse_frontmatter(
-            "---\nid: x\nsources:\n  - id: a\n    resource: file://x\n---\n\nbody\n"
+            "---\nid: x\nsources:\n  - id: a\n    uri: file://x\n---\n\nbody\n"
         )
 
 
 @pytest.mark.parametrize("fm", [
-    "generated: { by: x, at: y }\n", "sources: [{ id: a }]\n", "tags: [a, b]\n",
+    # The flow row carries a key OFF the G3 allowlist: the shape is admitted, the
+    # key is not, so this stays a T2 rejection and the door A/B half still holds.
+    # The two `sources` rows that lived here until 1.3.0 now PARSE - that is the
+    # G30 fix, not a weakening of this property. Their replacements are the same
+    # carriers with a key off the allowlist, so the row still measures what it
+    # says: the shape is admitted, the key set is not.
+    # The `tags` row moved the same way in 1.4.0: `[a, b]` now PARSES (P1), so
+    # the row carries the quoted form, which is still a T2 rejection.
+    "generated: { by: x, tool: y }\n", "sources: [{ id: a, uri: u }]\n", "tags: ['a', 'b']\n",
     "generated:\n  by: x\n", "generated.by: x\n",
-    "sources:\n  - id: a\n    resource: file://x\n",
+    "sources:\n  - id: a\n    uri: file://x\n",
+    "executor: [{ resource: skills/run.md }]\n",
 ])
 def test_t2_constrains_import_not_emission(fm):
     # T2 runs on door C only. The same frontmatter that FAIL_SECUREs through
@@ -738,3 +778,489 @@ def test_link_extraction_survives_the_redos_fix():
     assert extract_link_targets("see [x](./a.md) and [y](/b.md)") == ["./a.md", "/b.md"]
     assert extract_link_targets("[a b](./c.md)") == ["./c.md"]
     assert extract_link_targets("text [![img](./i.png)](./t.md)") == ["./i.png"]
+
+
+# --- G3: the typed, allowlisted mapping form (2026-08-21) --------------------
+# Door 1 of three (operator decision, 2026-08-21). The mapping *class* had no
+# expressible form, and OKF v0.2 writes its whole trust and provenance layer as
+# mappings — SPEC.md @ 62432a09 §5.2 uses flow form in its own examples, and §11
+# carries a hard MUST that presupposes they parse ("consumers MUST treat a bare
+# `verified` mapping as a one-element list"). A consumer measured 0 of 53
+# upstream concepts through the gate. This admits ONE shape: a flow mapping whose
+# every key is on the allowlist and whose every leaf is a plain scalar.
+
+def test_spec_flow_mapping_parses_into_a_typed_mapping():
+    # SPEC.md §5.2, verbatim. This is the red test: it must fail before the form
+    # exists and pass after, with a real dict — never a degraded string.
+    doc = (
+        "---\ntype: table\n"
+        "generated: { by: reference_agent/gemini-2.5-pro, at: 2026-06-20T22:53:05Z }\n"
+        "---\nbody\n"
+    )
+    assert parse_frontmatter(doc)[0]["generated"] == {
+        "by": "reference_agent/gemini-2.5-pro",
+        "at": "2026-06-20T22:53:05Z",
+    }
+
+
+def test_spec_bare_verified_mapping_parses():
+    # SPEC.md §5.2's bare form, which §11 turns into a hard MUST for consumers
+    # ("MUST treat a bare `verified` mapping as a one-element list") - a rule that
+    # cannot be obeyed by a consumer that cannot parse the mapping.
+    doc = "---\ntype: table\nverified: { by: human:ahormati, at: 2026-06-25T09:00:00Z }\n---\nb\n"
+    assert parse_frontmatter(doc)[0]["verified"] == {
+        "by": "human:ahormati", "at": "2026-06-25T09:00:00Z"}
+
+
+def test_spec_verified_list_of_flow_mappings_parses():
+    # §5.2's list form. This is the SAME typed form in list position, not the
+    # block-sequence-with-one-key route (`- uri: x`), which stays shut below.
+    doc = (
+        "---\ntype: table\nverified:\n"
+        "  - { by: human:ahormati, at: 2026-06-25T09:00:00Z }\n"
+        "  - { by: process:finance-nightly, at: 2026-06-26T02:00:00Z }\n"
+        "---\nbody\n"
+    )
+    assert parse_frontmatter(doc)[0]["verified"] == [
+        {"by": "human:ahormati", "at": "2026-06-25T09:00:00Z"},
+        {"by": "process:finance-nightly", "at": "2026-06-26T02:00:00Z"},
+    ]
+
+
+def test_spec_usage_window_parses():
+    doc = "---\ntype: table\nusage_window: { from: 2026-06-01T00:00:00Z, to: 2026-06-30T00:00:00Z }\n---\nb\n"
+    assert parse_frontmatter(doc)[0]["usage_window"] == {
+        "from": "2026-06-01T00:00:00Z", "to": "2026-06-30T00:00:00Z"}
+
+
+def test_an_unknown_key_inside_a_mapping_is_still_rejected():
+    # The rejection side of the allowlist. Without this test the allowlist could
+    # silently grow to "anything" - or be emptied - and nothing would fail.
+    with pytest.raises(OKFFrontmatterError) as exc:
+        parse_frontmatter("---\nid: x\ngenerated: { by: a, tool: shell }\n---\n\nbody\n")
+    assert "allowlist" in str(exc.value)
+
+
+def test_the_allowlist_is_not_empty_and_admits_only_the_spec_keys():
+    # Both directions of the same guard: a shrunk allowlist breaks the first
+    # assertion, a widened one the second.
+    for key in ("by", "at", "from", "to", "id", "title", "author", "usage_count",
+                "last_modified"):
+        assert parse_frontmatter(f"---\nid: x\nk: {{ {key}: v }}\n---\n\nb\n")[0]["k"] == {key: "v"}
+    for key in ("resource", "executor", "attester", "runtime", "command", "uri"):
+        with pytest.raises(OKFFrontmatterError):
+            parse_frontmatter(f"---\nid: x\nk: {{ {key}: v }}\n---\n\nb\n")
+
+
+_FLOW_REJECTED = [
+    # (id, value, what PyYAML 6.0.3 makes of it)
+    ("nested mapping", "{ by: { at: x } }", "a nested mapping"),
+    ("nested sequence", "{ by: [a, b] }", "a sequence leaf"),
+    ("anchor leaf", "{ by: &a x }", "an anchor definition, silently"),
+    ("tag leaf", "{ by: !!python/object:os.system x }", "refused outright"),
+    ("block scalar leaf", "{ by: | }", "a scanner error"),
+    ("nested colon leaf", "{ by: sub: v }", "refused outright"),
+    ("no space after colon", "{by:x}", "the KEY 'by:x', not a scalar"),
+    ("quoted leaf", "{ title: 'a, b' }", "a scalar - we refuse, deliberately"),
+    ("empty mapping", "{}", "an empty mapping"),
+    ("empty leaf", "{ by: }", "None"),
+    ("unclosed", "{ by: x", "a parse error"),
+    ("trailing junk", "{ by: x } more", "a parse error"),
+    ("duplicate key", "{ by: a, by: b }", "last-wins, silently"),
+]
+
+
+@pytest.mark.parametrize("cid,value,yaml_reads_as", _FLOW_REJECTED,
+                         ids=[c[0] for c in _FLOW_REJECTED])
+def test_the_mapping_form_admits_scalar_leaves_on_allowlisted_keys_only(cid, value, yaml_reads_as):
+    with pytest.raises(OKFFrontmatterError):
+        parse_frontmatter(f"---\nid: x\ngenerated: {value}\n---\n\nbody\n")
+
+
+def test_a_rejected_mapping_never_degrades_into_a_string():
+    # The 1.1.0 defect, re-asserted against the NEW form: a refused mapping must
+    # raise, not arrive as a str a consumer will .get() a key out of.
+    for value in ("{ by: { at: x } }", "{ tool: shell }", "{ by: x"):
+        with pytest.raises(OKFFrontmatterError):
+            parse_frontmatter(f"---\nid: x\ngenerated: {value}\n---\n\nbody\n")
+
+
+def test_the_admitted_mapping_is_a_dict_not_a_string():
+    value = parse_frontmatter("---\nid: x\ngenerated: { by: a, at: b }\n---\n\nb\n")[0]["generated"]
+    assert isinstance(value, dict), "a typed form that arrives as a str is the 1.1.0 defect"
+
+
+@pytest.mark.parametrize("cid,fm", [
+    ("block sequence, one key", "attester:\n  - resource: attesters/sql_equality.py\n"),
+    ("inline second colon", "attester: resource: attesters/sql_equality.py\n"),
+    ("block mapping", "attester:\n  resource: attesters/sql_equality.py\n"),
+    ("flow mapping, pointer key", "attester: { resource: attesters/sql_equality.py }\n"),
+])
+def test_the_pointer_routes_stay_shut(cid, fm):
+    # G3 is additive: none of the routes that put an executable-code pointer in a
+    # key the https allowlist never inspects is reopened. The fourth row is why
+    # `resource` is off the allowlist - the form would otherwise have carried the
+    # door-C pointer through in typed clothes instead of degraded ones.
+    doc = f"---\nid: x\ntype: Attested Computation\n{fm}---\n\nbody\n"
+    with pytest.raises(OKFFrontmatterError):
+        parse_frontmatter(doc)
+    assert import_bundle({"computations/x.md": doc}).disposition is Disposition.FAIL_SECURE
+
+
+def test_injection_in_a_mapping_leaf_is_caught_by_the_scan():
+    # T1 is not weakened by the new shape: a mapping leaf is scanned exactly like a
+    # scalar value or a list item. A typed form that parses but is not scanned would
+    # be a hole, not a fix.
+    doc = f"---\ntype: table\ngenerated: {{ by: {_INJECTION} }}\n---\nclean body\n"
+    assert scan_concept(doc).found is True
+
+
+def test_injection_in_a_listed_mapping_leaf_is_caught_by_the_scan():
+    doc = f"---\ntype: table\nverified:\n  - {{ by: {_INJECTION} }}\n---\nclean body\n"
+    assert scan_concept(doc).found is True
+
+
+def test_a_conformant_v02_trust_layer_now_reaches_the_gate():
+    # The measured consequence: a consumer reported 0 of 53 upstream concepts through
+    # the gate, because every one of them carries §5.2 trust frontmatter.
+    doc = (
+        "---\n"
+        "type: table\n"
+        "title: Users\n"
+        "resource: https://example.com/users\n"
+        "generated: { by: reference_agent/gemini-2.5-pro, at: 2026-06-20T22:53:05Z }\n"
+        "verified: { by: human:ahormati, at: 2026-06-25T09:00:00Z }\n"
+        "usage_window: { from: 2026-06-01T00:00:00Z, to: 2026-06-30T00:00:00Z }\n"
+        "---\nThe users table.\n"
+    )
+    result = import_bundle({"tables/users.md": doc})
+    assert result.disposition is Disposition.WARN
+    assert result.concepts[0].error is None
+
+
+# --- G30: the `sources` provenance layer becomes reachable (2026-09-02) ------
+# Door 2. G3 gave the mapping *class* one expressible form but left `sources`
+# unreachable: SPEC.md §5.1 writes an entry as a MAPPING carrying a REQUIRED
+# `resource`, so neither of the two carriers the spec and the producers actually
+# use could parse. Measured 02.09 by two consumers independently -- a flow
+# sequence of flow mappings raised on the `[` indicator, a block sequence of
+# block mappings raised "nested mappings are not supported".
+#
+# Why `resource` is admissible now when 1.2.0 argued it was not: the old
+# argument was that the parser could not tell `sources[].resource` (§5.1, a
+# citation) from `executor.resource` (§10, a code pointer). That premise was
+# false -- the owning key is in scope at every call site, it was simply never
+# threaded through. `resource` is allowlisted for `sources` entries ONLY, so
+# the door-C routes above stay shut on the same input.
+
+_SPEC_51_BLOCK = (
+    "sources:\n"
+    "  - id: ga4-schema\n"
+    "    resource: https://developers.google.com/analytics/bigquery/export-schema\n"
+    "    title: GA4 BigQuery Export schema\n"
+    "    author: team:ga4-docs\n"
+    "    usage_count: 5000\n"
+    "    last_modified: 2026-05-30T00:00:00Z\n"
+)
+
+
+def test_spec_sources_block_sequence_of_block_mappings_parses():
+    # SPEC.md §5.1's own example block, verbatim. It is the spec's canonical
+    # carrier for a REQUIRED field, so §11.1 ("parseable YAML frontmatter") makes
+    # a bundle written this way conformant -- refusing it refuses a conformant
+    # bundle, which is the failure mode G3 was opened to end.
+    fm, _ = parse_frontmatter(f"---\ntype: table\n{_SPEC_51_BLOCK}---\n\nbody\n")
+    assert fm["sources"] == [{
+        "id": "ga4-schema",
+        "resource": "https://developers.google.com/analytics/bigquery/export-schema",
+        "title": "GA4 BigQuery Export schema",
+        "author": "team:ga4-docs",
+        "usage_count": "5000",
+        "last_modified": "2026-05-30T00:00:00Z",
+    }]
+
+
+@pytest.mark.parametrize("cid,fm,expected", [
+    ("one entry",
+     "sources: [{ id: golden-v0-2-sales, resource: fixture }]\n",
+     [{"id": "golden-v0-2-sales", "resource": "fixture"}]),
+    ("two entries",
+     "sources: [{ id: a, resource: https://e.com/a }, { id: b, resource: https://e.com/b }]\n",
+     [{"id": "a", "resource": "https://e.com/a"},
+      {"id": "b", "resource": "https://e.com/b"}]),
+])
+def test_sources_flow_sequence_of_flow_mappings_parses(cid, fm, expected):
+    # The form the producer emits today (llm-ingestion-okf's golden
+    # expected-bundle/ingest-sales.md, measured 02.09). One entry raised too, so
+    # this is not an arity bug: the `[` indicator refused the carrier outright.
+    assert parse_frontmatter(f"---\ntype: table\n{fm}---\n\nbody\n")[0]["sources"] == expected
+
+
+def test_the_two_sources_carriers_parse_to_the_same_value():
+    block = parse_frontmatter(
+        "---\ntype: t\nsources:\n  - id: a\n    resource: https://e.com/a\n---\n\nb\n")[0]
+    flow = parse_frontmatter(
+        "---\ntype: t\nsources: [{ id: a, resource: https://e.com/a }]\n---\n\nb\n")[0]
+    assert block["sources"] == flow["sources"] == [{"id": "a", "resource": "https://e.com/a"}]
+
+
+@pytest.mark.parametrize("cid,fm", [
+    ("unknown key, flow", "sources: [{ id: a, uri: https://e.com/a }]\n"),
+    ("unknown key, block", "sources:\n  - id: a\n    uri: https://e.com/a\n"),
+    ("unknown key, flow value", "sources: { id: a, kind: doc }\n"),
+])
+def test_an_unknown_key_in_a_sources_entry_is_still_rejected(cid, fm):
+    # The negative control. The carrier is admitted; the key set is not. A
+    # producer-invented key gets no free ride on the new shape.
+    with pytest.raises(OKFFrontmatterError) as exc:
+        parse_frontmatter(f"---\ntype: table\n{fm}---\n\nbody\n")
+    assert "allowlist" in str(exc.value)
+
+
+@pytest.mark.parametrize("cid,fm", [
+    ("executor, block sequence", "executor:\n  - resource: skills/run-on-bq.md\n"),
+    ("executor, flow sequence", "executor: [{ resource: skills/run-on-bq.md }]\n"),
+    ("attester, flow sequence", "attester: [{ resource: attesters/sql_equality.py }]\n"),
+    ("attester, block sequence", "attester:\n  - resource: attesters/sql_equality.py\n"),
+])
+def test_resource_is_allowlisted_for_sources_entries_only(cid, fm):
+    # The whole reason `resource` can be admitted at all: the owning key decides.
+    # §10's `executor.resource` / `attester.resource` name run instructions and
+    # code -- door C -- and stay refused through EVERY carrier, including the two
+    # this change opens for `sources`.
+    doc = f"---\nid: x\ntype: Attested Computation\n{fm}---\n\nbody\n"
+    with pytest.raises(OKFFrontmatterError) as exc:
+        parse_frontmatter(doc)
+    assert "allowlist" in str(exc.value)
+    assert import_bundle({"computations/x.md": doc}).disposition is Disposition.FAIL_SECURE
+
+
+@pytest.mark.parametrize("cid,fm", [
+    # SPEC §5.1: "A single entry MAY carry its own `usage_window`". That is a
+    # mapping inside a mapping -- depth 2 -- and stays refused: no nesting deeper
+    # than one is a security property this change does not spend. Registered as a
+    # conformance gap in docs/LIMITATIONS.md, not as an oversight.
+    ("per-entry usage_window, flow",
+     "sources: [{ id: a, usage_window: { from: x, to: y } }]\n"),
+    ("per-entry usage_window, block",
+     "sources:\n  - id: a\n    usage_window: { from: x, to: y }\n"),
+    # A sequence inside an entry is the same depth violation.
+    ("nested sequence", "sources: [{ id: a, title: [x, y] }]\n"),
+])
+def test_a_sources_entry_admits_scalar_leaves_only(cid, fm):
+    with pytest.raises(OKFFrontmatterError):
+        parse_frontmatter(f"---\ntype: table\n{fm}---\n\nbody\n")
+
+
+@pytest.mark.parametrize("cid,fm", [
+    ("scalar then mapping", "sources:\n  - https://e.com/a\n  - id: b\n    title: B\n"),
+    ("mapping then scalar", "sources:\n  - id: a\n    title: A\n  - https://e.com/b\n"),
+])
+def test_a_block_list_may_not_mix_scalars_and_mappings(cid, fm):
+    # A consumer that reads `entry.get("id")` over the list crashes on the str.
+    # One list, one item type -- refuse rather than hand back a mixed tree.
+    with pytest.raises(OKFFrontmatterError) as exc:
+        parse_frontmatter(f"---\ntype: table\n{fm}---\n\nbody\n")
+    assert "mix" in str(exc.value)
+
+
+@pytest.mark.parametrize("cid,fm", [
+    ("empty flow sequence", "sources: []\n"),
+    ("flow sequence, unclosed", "sources: [{ id: a }\n"),
+    ("flow sequence, trailing junk", "sources: [{ id: a }] x\n"),
+    ("flow sequence, nested sequence", "sources: [[ id ]]\n"),
+])
+def test_the_flow_sequence_refuses_a_carrier_it_cannot_read(cid, fm):
+    # What no element shape rescues: nothing at all, an unclosed mapping, junk
+    # between elements, and a sequence inside a sequence. The plain-scalar
+    # element admitted in 1.4.0 (P1) is a THIRD shape, not a loosening of these
+    # -- each row here still raises through it.
+    with pytest.raises(OKFFrontmatterError):
+        parse_frontmatter(f"---\ntype: table\n{fm}---\n\nbody\n")
+
+
+# --- P1: the flow sequence of plain scalars (operator decision, 2026-09-08) --
+# `tags: [a, b, c]` is the ONE form SPEC.md §4.1's own skeleton writes out for
+# `tags`, and the only candidate measured to move the upstream corpus at all:
+# 0/53 -> 6/53 against the pinned reference bundles (`_okf-upstream` @ 3fcbb9f,
+# denominator 53). The two neighbours measured 0/53 EACH and are deliberately
+# NOT built -- P2 (a flush-left block sequence) and P3 (a folded plain scalar)
+# would cost parser surface and buy no measured conformance on this corpus;
+# see docs/2026-09-07-limitations-44-maaling.md §§7-8.
+#
+# It does not spend the depth-1 rule: the elements are leaves, the same depth
+# the block list of scalars already carries. The binding constraint on the
+# corpus is elsewhere and stays refused -- 44/53 stop on `generated` written as
+# a top-level block mapping, which is a security decision, not this one.
+
+@pytest.mark.parametrize("cid,fm,expected", [
+    ("SPEC §4.1 skeleton", "tags: [finance, revenue, headline-metric]\n",
+     ["finance", "revenue", "headline-metric"]),
+    ("two elements", "tags: [a, b]\n", ["a", "b"]),
+    ("one element", "tags: [solo]\n", ["solo"]),
+    ("uneven spacing", "tags: [ a ,b ]\n", ["a", "b"]),
+    ("one trailing comma, as the flow mapping already allows", "tags: [a, b,]\n", ["a", "b"]),
+])
+def test_a_flow_sequence_of_plain_scalars_parses(cid, fm, expected):
+    # The red test for P1: `[a, b]` must come back as a real list of strings,
+    # never a degraded string (the 1.1.0 defect) and never a refusal.
+    assert parse_frontmatter(f"---\ntype: table\n{fm}---\n\nbody\n")[0]["tags"] == expected
+
+
+def test_the_two_scalar_sequence_carriers_parse_to_the_same_value():
+    flow = parse_frontmatter("---\ntype: t\ntags: [a, b]\n---\n\nb\n")[0]
+    block = parse_frontmatter("---\ntype: t\ntags:\n  - a\n  - b\n---\n\nb\n")[0]
+    assert flow["tags"] == block["tags"] == ["a", "b"]
+
+
+@pytest.mark.parametrize("cid,fm", [
+    # Quoting is the failure mode this shape is refused for elsewhere: the
+    # parser retains quotes rather than stripping them, so admitting a quoted
+    # element would hand back a value YAML reads differently.
+    ("single-quoted element", "tags: ['a']\n"),
+    ("double-quoted element", 'tags: ["a", "b"]\n'),
+    # Comma splitting is character-level, so anything that would need YAML
+    # semantics to split correctly is refused rather than guessed at.
+    ("colon inside an element", "tags: [a: b]\n"),
+    ("comment indicator", "tags: [a #b]\n"),
+    ("brace inside an element", "tags: [a{b}]\n"),
+    ("bracket inside an element", "tags: [a[b]]\n"),
+    # Depth: a sequence inside a sequence opens a second level and is refused
+    # on the same character rule, with no YAML semantics involved.
+    ("nested flow sequence", "tags: [[a]]\n"),
+    ("empty element", "tags: [a, , b]\n"),
+    ("leading empty element", "tags: [, a]\n"),
+    ("empty sequence", "tags: []\n"),
+    ("unterminated sequence", "tags: [a, b\n"),
+    # The unchanged scalar indicators still apply to an element, exactly as they
+    # do to a block-list item: an anchor or an alias is not a plain scalar.
+    ("anchor element", "tags: [&anchor]\n"),
+    ("alias element", "tags: [*alias]\n"),
+    ("explicit tag element", "tags: [!!python/object]\n"),
+])
+def test_a_flow_sequence_scalar_element_must_be_a_plain_scalar(cid, fm):
+    with pytest.raises(OKFFrontmatterError):
+        parse_frontmatter(f"---\ntype: table\n{fm}---\n\nbody\n")
+
+
+@pytest.mark.parametrize("cid,fm", [
+    ("scalar then mapping", "tags: [a, {b: c}]\n"),
+    ("mapping then scalar", "sources: [{ id: a }, plain]\n"),
+])
+def test_a_flow_sequence_may_not_mix_scalars_and_mappings(cid, fm):
+    # Same rule, same reason as the block list: a consumer iterating the value
+    # and reading `entry.get("id")` crashes on the first str. The mixing verdict
+    # is reached BEFORE the element is parsed, so it is what the caller sees --
+    # not an allowlist complaint about a key that was never the problem.
+    with pytest.raises(OKFFrontmatterError) as exc:
+        parse_frontmatter(f"---\ntype: table\n{fm}---\n\nbody\n")
+    assert "mix" in str(exc.value)
+
+
+@pytest.mark.parametrize("cid,fm,expected", [
+    ("one entry", "sources: [{ id: x }]\n", [{"id": "x"}]),
+    ("two entries", "sources: [{ id: a, title: A }, { id: b, title: B }]\n",
+     [{"id": "a", "title": "A"}, {"id": "b", "title": "B"}]),
+])
+def test_the_g30_flow_mapping_carrier_is_unchanged_by_the_scalar_element(cid, fm, expected):
+    # The no-regression pin. P1 adds a branch to the same function that carries
+    # `sources`; the mapping element must parse exactly as it did in 1.3.0.
+    assert parse_frontmatter(f"---\ntype: table\n{fm}---\n\nbody\n")[0]["sources"] == expected
+
+
+# The self-safety row for P1 (OWASP LLM10). The predicate is character-level
+# with no regex, so there is no backtracking engine to blow up -- but "no regex"
+# is an argument, and the sweep in docs/redos-sweep.py cannot check it, because
+# it collects compiled patterns and this predicate compiles none. So it is
+# measured instead, on the same CPU clock every other bound here uses. Over four
+# doublings (50k -> 800k, 2026-09-08) the exponent is 0.86-0.99 in the element
+# LENGTH and 0.97-1.05 in the element COUNT: linear in both, 0.66s at 800_000
+# elements and 0.03s at an 800_000-character element. The refusal path is the
+# same shape -- a forbidden character at the very end of an 800_000-character
+# element is found in 0.04s -- because the element is scanned once per rule, not
+# rescanned per start position.
+_FLOW_SCALAR_PERF_N = 400_000
+
+
+def test_a_long_flow_sequence_of_scalars_stays_bounded():
+    many = "---\ntype: t\ntags: [" + ", ".join(["ab"] * _FLOW_SCALAR_PERF_N) + "]\n---\nbody\n"
+    assert scan_seconds(parse_frontmatter, many) < 2.0
+    one_long = "---\ntype: t\ntags: [" + "a" * _FLOW_SCALAR_PERF_N + "]\n---\nbody\n"
+    assert scan_seconds(parse_frontmatter, one_long) < 2.0
+
+
+def test_a_long_flow_sequence_element_is_refused_without_a_rescan():
+    # The refusal is the half an attacker controls: a forbidden character parked
+    # at the END of a long element is the worst case for any per-start rescan.
+    doc = "---\ntype: t\ntags: [" + "a" * _FLOW_SCALAR_PERF_N + ":]\n---\nbody\n"
+
+    def refuse(payload):
+        with pytest.raises(OKFFrontmatterError):
+            parse_frontmatter(payload)
+
+    assert scan_seconds(refuse, doc) < 2.0
+
+
+def test_injection_in_a_sources_entry_leaf_is_caught_by_the_scan():
+    # T1 over the new shape: every leaf of every entry reaches scan_output.
+    doc = f"---\ntype: table\nsources: [{{ id: a, title: {_INJECTION} }}]\n---\nclean\n"
+    assert scan_concept(doc).found is True
+    doc_block = f"---\ntype: table\nsources:\n  - id: a\n    title: {_INJECTION}\n---\nclean\n"
+    assert scan_concept(doc_block).found is True
+
+
+def test_the_producer_golden_now_passes_the_gate():
+    # llm-ingestion-okf's expected-bundle/ingest-sales.md, the K5 blocker.
+    doc = (
+        "---\n"
+        "type: dataset\n"
+        "title: Regional Sales\n"
+        "source_system: golden-v0-2-sales\n"
+        "ingested_at: 2026-07-16T12:00:00Z\n"
+        "generated: { by: process:okf-ingest, at: 2026-07-16T12:00:00Z }\n"
+        "sources: [{ id: golden-v0-2-sales, resource: fixture }]\n"
+        "---\n\n| region | units |\n| --- | --- |\n| nord | 412 |\n"
+    )
+    result = import_bundle({"datasets/sales.md": doc})
+    assert result.disposition is Disposition.WARN
+    assert result.concepts[0].error is None
+
+
+@pytest.mark.parametrize("cid,fm", [
+    # A `- ` item whose text begins with a YAML indicator AND carries a `": "`
+    # must fall through to the scalar predicates, not into the block-mapping
+    # route with a half-validated key. Each of these fails `_KEY_RE` on the key
+    # side, so `_reject_dangerous_value` gets the item intact.
+    ("anchor", "sources:\n  - &anchor id: a\n"),
+    ("alias", "sources:\n  - *anchor id: a\n"),
+    ("tag", "sources:\n  - !!str id: a\n"),
+    ("directive", "sources:\n  - %YAML id: a\n"),
+    ("reserved", "sources:\n  - `x id: a\n"),
+    ("merge key", "sources:\n  - id: a\n    <<: *base\n"),
+])
+def test_an_indicator_in_a_block_item_is_refused_before_the_mapping_route(cid, fm):
+    with pytest.raises(OKFFrontmatterError):
+        parse_frontmatter(f"---\ntype: t\n{fm}---\n\nbody\n")
+
+
+@pytest.mark.parametrize("cid,fm", [
+    # The block mapping ends at a blank line and at a line in column zero. Both
+    # hand control back with an index that must not skip or re-read a line: a
+    # dangling `resource:` line left over from a mapping that closed early must
+    # RAISE, never be silently dropped -- a pointer that vanishes rather than
+    # failing is exactly this repo's failure class.
+    ("blank line inside the mapping", "sources:\n  - id: a\n\n    resource: b\n"),
+    ("top-level key interleaved", "sources:\n  - id: a\ntags: x\n    resource: b\n"),
+])
+def test_a_line_orphaned_by_the_mapping_boundary_raises_rather_than_vanishing(cid, fm):
+    with pytest.raises(OKFFrontmatterError):
+        parse_frontmatter(f"---\ntype: t\n{fm}---\n\nbody\n")
+
+
+def test_the_block_list_hands_back_an_index_that_resumes_at_the_next_key():
+    # The return-index contract: a top-level key following a multi-entry block
+    # list is neither swallowed by the list nor re-read as a list item.
+    fm, _ = parse_frontmatter(
+        "---\ntype: t\nsources:\n  - id: a\n    resource: b\n  - id: c\ntitle: T\n---\n\nbody\n")
+    assert fm == {"type": "t", "sources": [{"id": "a", "resource": "b"}, {"id": "c"}],
+                  "title": "T"}