# Changelog

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).

Versioning note: the repository tag versions **the contract** (file set, key names,
case ids, disposition semantics). Each JSON file additionally carries its own
`"version"` field, bumped when that file changes.

## [Unreleased]

Nothing yet.

## [0.3.0] — 2026-08-11

**A detection pattern changed value. That has not happened in this repository before, and it
is the reason this is a release.** `v0.2.0` changed what a runtime must *declare*; this one
changes what a conforming runtime *finds*. A consumer that vendors `0.3.0` and re-runs its
suite will see a finding on inputs that produced none under `0.2.0`. In 0.x that is a minor
bump by the rules; read the first entry below before upgrading, not the version number.

The lexicon `0.6.0` entry previously sitting under *Unreleased* is folded in here — it was
committed but never tagged, and `0.7.0` supersedes its central claim.

### Changed

- **`lexicon/injection-lexicon.json` (`0.5.1` → `0.7.0`) — `hybrid-xss:script-tag` converged
  on `llm-security`'s current form.**

  ```
  0.6.0 and earlier   <script\b[^>]*>[\s\S]*?<\/script>     closing tag REQUIRED
  0.7.0               <script\b[^>]*>                       opening tag only
  ```

  Byte-identical to `llm-security`'s `RegExp.prototype.source` at `b1ba1fb`
  (`scanners/lib/injection-patterns.mjs:170`), verified by compiling both. They dropped the
  closing-tag requirement in `90f576f` (2026-08-10) because it was a recall hole:
  `<script>alert(1)` and `<script src=x.js>` both returned no finding.

  **This is re-extraction, not revision, and the distinction is the whole justification.**
  This repository's behaviour-preservation invariant forbids commons from *correcting* seed
  data it believes is wrong — that rule stands and was not weakened. It does not forbid
  re-reading the source after the source itself moved and its owner asked. The lexicon's
  declared provenance is `llm-security`'s injection table, and being loadable verbatim by
  that table's owner is the one thing it exists to do; the standing alternative was a
  per-pattern override in `llm-security`'s own loader, i.e. a published core its source
  repository could not load as published.

  **Direction matters for what it cannot break:** the new form matches a strict superset of
  the old one, so relative to `0.6.0` it can add matches and cannot introduce a false
  negative. The reverse change would not have been adoptable on the same reasoning.

  Operator decision, 2026-08-11, on `llm-security`'s blocking request. Explicitly **not**
  decided by the 2-of-3 majority across the three ports: a count of implementations is not a
  mandate over detection data, and the provenance argument would hold with the guard on
  either side.

  Measured collateral: **none.** The full corpus was run under both patterns — 84/84
  lexicon-scoped cases pass under `0.7.0`, and exactly one case's finding set differs between
  the two forms (the new one below). The widening added no finding to any other case's input.

  `source_fidelity` restructured to keep its numbers coordinate-bearing:
  `patterns_byte_identical_to_source` keeps its key and its value (83) and gains the field it
  was missing, `byte_identical_against_commit: b1ba1fb`. Against the original extraction
  commit `b0de0ca` this file is now 82/83, recorded as `count_against_extraction_commit`.
  `post_extraction_drift` — added in the folded-in `0.6.0` to record the then-open divergence
  — is now marked `status: resolved in 0.7.0 by re-extraction` and carries the before/after
  pattern text, so a consumer diffing against either commit has a coordinate for what it
  finds.

- **`conformance/manifest.json` (`0.2.0` → `0.3.0`) — `case_id_derivation` extended with an
  optional variant suffix.**

  ```
  before   case_id = <pattern_id, ':' → '__'>
  after    case_id = <pattern_id, ':' → '__'>  [ '--' <variant-slug> ]
  reverse  truncate at first '--', then '__' → ':'
  ```

  No existing case id moved, so this is additive. `--` was measured absent from all 83
  ratified pattern ids and all 89 pre-existing case ids, which keeps the reverse transform
  purely lexical — no lookup against the id list — the property the original one-to-one rule
  was protecting.

  The `one_case_per_pattern_id` key is **removed**, superseded by
  `case_id_derivation.variant_suffix.supersedes`, which quotes its text. It was documentation
  of the constraint, not data a consumer matches on, but it is called out here because a
  removed key is normally a breaking change in this repository.

  `omitted_payloads` gains `derivation_ground_withdrawn_in_0_3_0`: the guard's seventh
  active-content payload was omitted on two grounds, and this change retires one of them. The
  other stands, so the payload stays omitted — on one ground instead of two. **It was not
  added back**; that is a separate decision, not a consequence of this one.

### Added

- **`conformance/hybrid-xss__script-tag--src-no-close/` (89 → 90 cases)** — input
  `<script src=x.js>`, 17 bytes, expecting `hybrid-xss:script-tag`. The regression gate for
  the change above, and the reason the corpus could not previously see it: the existing
  `hybrid-xss__script-tag` input `<script>steal()</script>` matches the pattern under *both*
  forms, so it passes either way. Reverting the pattern to its `0.6.0` form fails this case
  and only this case — mutation-verified in both directions across all 90.

  **The first case input authored in this repository** rather than reproduced verbatim from a
  runtime's payload set, recorded in the new `authored_payloads` block rather than folded into
  `payload_provenance`, whose value is precisely the claim that its inputs are verbatim
  upstream. That claim stays exactly as strong as it was: 83 of 83. Both witnesses for this
  axis were named by `llm-security` on 2026-08-10; this is the first of the two. Findings
  measured through the guard's public API at `0dce50f` / `0.5.0`, with the existing case's
  committed bytes and digest reproduced by the same harness in the same run as a control.

- **`schema/conformance-declaration.schema.json` (`0.1.0`)** — the shape a runtime publishes
  alongside a conformance result, satisfying the §1.1 MUST that `v0.2.0` created and left
  without a form. Requested by `llm-security` in those terms (runtime, commit measured,
  implemented file paths) with the stated reason that two runtimes publishing free-form
  declarations makes `83/83 + 6 not-applicable` unparseable by anyone but its author.

  Carries the two arithmetic invariants §1.1 implies but cannot state unambiguously in prose:
  the four verdict counts MUST sum to the total, and the total MUST equal the corpus case
  count at the commit measured. Requires the enumeration arrays whenever their counts are
  non-zero, which turns §1.1's "MUST still be enumerated" from prose into a schema failure.
  Keeps `error` and `not_applicable` structurally distinct, per §1.1. Records
  `declaration_source` — whether the declared set is derived from the runner's own constant or
  hand-maintained beside it — because only the derived form makes the anti-narrowing fence
  structural. **Deliberately not a gate:** nothing in this repository runs, and no validation
  was asked for. Mutation-tested: the example validates, and five distinct defect classes are
  rejected.

- **`spec/conformance-corpus.md` §1.1** — normative pointer to that schema, plus a SHOULD that
  a runtime derive its declared set from the constant its runner uses to accept or reject a
  `scope`, and record which it did.

### Fixed

- `docs/lexicon-port-divergence.md` — the `hybrid-xss:script-tag` row is closed on the
  closing-tag axis, having reversed twice in three days (guard-diverges → commons-diverges →
  converged). What remains open is the one-byte span difference: the guard excludes `<` from
  its negated class and the other two do not, so `<script <x>` matches commons and
  `llm-security` and not the guard. Measured by compiling all three forms, not reasoned from
  the character classes; neither side has claimed it.

## [0.2.0] — 2026-08-11

The contract gained a normative MUST, which is why this is a release rather than a
metadata commit: **a runtime that conformed to `v0.1.0` does not conform to this one until
it declares the set of commons data files it implements.** In 0.x that is a minor bump by
the rules, but it is breaking in substance, and a consumer reading only the version number
should learn that here rather than from a failing suite.

Everything below this heading was previously listed as unreleased.

### Retracted

- **The claim that the Python guard's port cites `severity.mjs` for hybrid severity.** It is
  false. It was carried in three places — `lexicon/injection-lexicon.json`
  (`families[hybrid].severity_provenance.not_from`), `docs/lexicon-port-divergence.md`
  (§ *Severity: the 8 hybrid patterns*), and the `[0.1.0]` entry below — and it was never
  measured here. It restated an assertion received from `llm-security` (coord message
  `20260809T201048Z`) as a commons finding.

  Measured against the guard's own tree, which `llm-ingestion-pipeline-security` asked for
  twice before this was checked: `severity.mjs` has **never** appeared in
  `src/llm_ingestion_guard/injection_lexicon.json` at any point in that file's history
  (`git log -S` returns no commits), and at `0bf0729` — the commit
  `conformance/manifest.json` pins — the only tree-wide occurrence is `docs/PLAN.md:114`,
  correctly attributing the *report* module to `output.mjs` + `severity.mjs`. The guard's
  only source statement for the lexicon is the `note` at `injection_lexicon.json:3`, and it
  names `injection-patterns.mjs`.

  **No detection data moves.** `families[hybrid].severity` is still `high`, still sourced to
  `injection-patterns.mjs:274-281`, re-verified at `b0de0ca`; `severity.mjs` still contains
  zero occurrences of `CRITICAL_PATTERNS`, `HIGH_PATTERNS`, `MEDIUM_PATTERNS` and
  `HYBRID_PATTERNS`, re-measured the same day. Only the sentence about the *other* repository
  falls.

  The retraction is marked in place rather than edited away, and it is worth naming why this
  one survived review: the claim arrived bundled with a correct measurement of the same
  question, from a repository that had done its half properly. The correct half carried the
  incorrect half past the check — which is precisely the defect
  `severity_provenance.not_from` was written to warn about, one level up.

### Added

- **`not-applicable`, a third conformance verdict** (`spec/conformance-corpus.md` §1.1). A
  runtime now declares the set of commons data files it implements; a case whose `scope`
  names a file outside that set is `not-applicable` rather than failed. §1 alone would have
  reported an architectural difference as a defect — one seeding runtime has no
  active-content table and never will, and 7 permanent failures say nothing a reader can use.
  The verdict is fenced so it cannot become an exit: it attaches to a **table**, never to a
  case (per-case opt-out is the silent skip §1 forbids), a declared set MUST NOT be narrowed
  to convert failures into `not-applicable`, and such cases MUST still be enumerated rather
  than dropped from the denominator. §8 now states the consequence: a pass count is
  unreadable without the declared set beside it.

- **Six active-content conformance cases** — `active__markdown-image`, `active__markdown-link`,
  `active__reference-link`, `active__autolink`, `active__raw-html`, `active__data-uri`. The
  corpus goes 83 → 89, and `scope_covered` gains `signatures/active-content.json`.

  Generated from measurement, not written: payload strings were extracted from the seed
  runtime's `coverage.py` by AST — evaluating each `_scan_case` argument in that module's own
  namespace rather than retyping detection data — then run through its public output gate.
  The fixtures were then re-read from disk by a separate checker that re-computed every
  digest, re-scanned the bytes and applied `exact-within-scope` independently of the
  generator, because a generator agreeing with itself proves nothing: 6 cases, 0 failed
  checks.

  **These six prove less than the 83, and the manifest says so.** Their payloads come from
  the only runtime implementing the table, so no second implementation's agreement could be
  measured. They pin one runtime's behaviour as a contract a future implementer can be held
  to — which is less than cross-runtime agreement and more than nothing.

- `signatures/active-content.json` **0.1.0 → 0.2.0** — a `pattern_id_space` block. Unlike the
  lexicon's, nothing was constructed: `label_format` and the `constructs` keys were already
  extracted verbatim, and their concatenation *is* what the seed runtime emits. The block
  states an id space the file already had implicitly, and records that it is ratified by
  **one** runtime rather than two.

### Changed

- `lexicon/injection-lexicon.json` **0.5.0 → 0.5.1** — provenance metadata only; no pattern,
  id, alias, family or severity value changes.

- `conformance/manifest.json` **0.1.1 → 0.2.0** — the six cases, `scope_covered`,
  `count_by_scope`, separate provenance and measurement blocks for the active-content half
  (a different source structure at a different commit; one pin must not stand for two
  measurements), and `scope_planned.blockers`.

- **`spec/conformance-corpus.md` §4 no longer claims scoping "asks a question both can
  answer."** That held only while every case was scoped to the one table both runtimes
  implement, and stopped being true the moment a case was scoped to a single-runtime table.
  Scope narrows *what* is compared; it does not make every runtime a valid addressee. The
  superseded sentence is named in place rather than edited away.

### Fixed

- **§4 now states that "belongs to a data file" means published there, never "shares its
  prefix."** The distinction has a live witness: the seed runtime emits `active:oversize-input`,
  a self-safety flag about its own scan cap, which carries the `active:` prefix but is no
  construct in `signatures/active-content.json`. A prefix-matching runtime would pull it into
  the comparison and fail a case over a finding the corpus makes no claim about. Recorded in
  that file under `pattern_id_space.not_every_active_label` as well.

- **§6 now states the derivation's cost.** `case_id` derives from `pattern_id` alone, so a
  single-finding scope holds at most one case per pattern id — there is nowhere in the name
  for a second. The seed runtime's matrix drives *two* payloads at `active:markdown-image`;
  measured, their in-scope finding sets are identical, and the second's only distinguishing
  signal (`entropy:base64-blob`) falls outside every table this repository publishes. It was
  dropped rather than given a discriminated id, which would have broken the reverse
  transform, and it is named in `conformance/manifest.json` under `omitted_payloads` so that
  6 built from 7 offered reads as a decision rather than a miscount.

### Measured, not shipped

- **The remaining four cases are blocked on two distinct unresolved questions**, now recorded
  under `scope_planned.blockers` instead of the earlier blanket "no runtime has agreed to an
  id space". That framing was wrong for both:

  - **Carriers (3).** No adoptable id space, and a second problem underneath. The guard emits
    two *stage-coupled* labels for one carrier — `sanitize:zero-width` on input,
    `output:zero-width-present` on output, same split for bidi and unicode-tag — while
    llm-security emits prose titles. A commons id must be invented stage-neutral, which no
    other id space here required. And since `exact-within-scope` compares a finding *set*, an
    id aliasing both labels makes the verdict depend on which entry point the runtime was
    measured through — an entry-point dependence the lexicon cases do not have.
  - **Secret egress (1).** Not an id-naming question at all. The two runtimes carry
    **different tables**: 18 entries here against the guard's 25, cut at different
    granularities (this file's single `GitHub Token` is four ids there, `Private Key PEM
    Block` three, `Database connection string` four), with membership diverging both ways.
    `aws-access-key-id` is the one clean 1:1 — which is why exactly one egress case was ever
    offered. That number was a symptom, not modesty. A shared id space presupposes a table
    reconciliation nobody has done.

## [0.1.0] — 2026-08-10

Initial extraction. Runtime-neutral detection data, the finding contract, and a conformance
corpus, extracted from the `llm-security` Node implementation and a Python guard **without
behaviour change** — that invariant is the release, not a caveat on it.

What the tag is worth resting on: seven of the eight JSON artefacts were rebuilt from the
commons file alone and diffed against their source implementation, three of them against the
source module at a pinned commit. The eighth says `verified: false` about itself. The corpus
holds 83 cases on which both seeding runtimes were measured agreeing exactly.

What it is not: `spec/decode-pipeline.md` does not exist, and the corpus constrains one of
the seven data files. Both absences are named in *Not included* rather than papered over.

### Added

- `conformance/` — **83 cases, one per injection-lexicon pattern**, plus `manifest.json`.
  Each case is a directory holding `input.txt` (the exact bytes, no trailing newline) and
  `expected.json` (the findings, named by commons pattern `id`).

  Both seeding runtimes were measured producing the **same lexicon finding set on all 83**,
  through their public entry points — `scanForInjection()` at `b0de0ca` and
  `scan_output(source=OUTPUT)` at `0bf0729` — with labels mapped to commons ids through the
  lexicon's own `aliases` block. Not through rebuilt regex tables: a table-level comparison
  yields a number that describes neither runtime, which is the mistake the divergence
  document had to retract.

  **The 13 divergent patterns are in, unmarked, and that is the substantive result.** Their
  divergence was measured on witness inputs — an attribute run padded past 256 characters,
  an interior `<`, an unclosed `<script>` — and none of those shapes occurs in a corpus
  payload. All 13 agree on their own case input. Nobody had to pick whose recall cost
  becomes the contract, because the question was never reachable from these inputs. A
  per-case caveat would have asserted a doubt the measurement disproves.

  Inputs are the guard's `coverage.py` payloads, reproduced verbatim. One runtime authored
  them; what makes them a cross-runtime corpus is the measurement through the other, and the
  manifest records the asymmetry rather than averaging it away.

- `spec/conformance-corpus.md` — **normative.** How a case is read: `input.txt` is bytes and
  is not to be trimmed or re-encoded, `expected.json` names findings by `pattern_id` only
  (severity and OWASP anchor are looked up in the lexicon, never restated), and
  `exact-within-scope` requires equality **restricted to the data files the case names**.

  The field is `pattern_id`, not `id`, because this repository already publishes an unrelated
  finding `id`: `schema/finding.schema.json` defines it as `DS-<scanner>-<counter>` from a
  process-global counter — stable across neither runs nor processes. Two normative documents
  using one word for a stable rule identity and a volatile per-emission sequence number would
  have produced runtimes failing every case for reasons unrelated to detection. §3.1 states
  the distinction and publishes the bridge a runtime actually needs: its own label maps to a
  `pattern_id` through the lexicon's `aliases` object, and a runtime absent from that object
  has no published way to be compared at all.

  Scoping is what makes exactness safe — the two runtimes do not implement the same set of
  tables, so a whole-report comparison would fail for reasons unrelated to the pattern under
  test. Exactness is what makes the corpus worth running — a contains-only corpus is passed
  by a runtime that flags everything. `observed_out_of_scope` is evidence, never expectation,
  and an absent runtime key means **unmeasured**, not measured-empty.

  The document also states the one place this repository's "every JSON file carries a
  top-level `version`" convention does not apply: fixtures are versioned as a corpus, in
  `conformance/manifest.json`. Stated rather than left to be discovered.

- `schema/finding.schema.json` — the finding contract plus the SARIF output profile.
  Normative. Closed against the producer in 0.2.0; the JSONL profile is `not applicable`.
- `signatures/active-content.json` — the EchoLeak class (CVE-2025-32711): 17 patterns,
  severities, opacity floors and pass order, from the Python guard.
- `lexicon/injection-lexicon.json` — 83 prompt-injection patterns in four families
  (21 critical, 32 high, 22 medium, 8 hybrid).
- `codepoints/carriers.json` — six carrier tables: zero-width characters, the Unicode Tags
  block, the Supplementary Private Use Areas, BIDI controls, the Cyrillic presence set and
  the 28-entry fold-to-Latin homoglyph map.
- `signatures/secret-egress.json` — the 18 fixed credential and token shapes. Array order
  is normative.
- `mapping/owasp-map.json` — four taxonomy maps (LLM, ASI, AST, MCP) over one shared
  16-prefix key set.
- `calibration/calibration.json` — risk-score tier constants, verdict thresholds, risk-band
  cutoffs, posture grade thresholds.
- `signatures/malware-signatures.json` — the known-bad-identity table for the `SIG` class:
  seven signatures over four families (`webshell`, `reverse_shell`, `cryptominer`,
  `hacktool`), reproduced verbatim from `knowledge/signatures.json` at `b0de0ca`, key order
  included, with the source file's byte length and SHA-256 pinned in `provenance`.

  The rules were the easy half. The file's substance is the line between the table and the
  engine, drawn in `engine_behaviour_not_data`: **no rule carries a `flags` field**, because
  the engine compiles every pattern with `i` unconditionally — so a consumer that compiles
  these case-sensitively silently under-matches all seven. Each pattern is also run against
  five decode variants, not just raw bytes; rules are filtered by an enabled-families policy;
  a rule fires once per file; operator rules are merged at scan time; and the loader defaults
  four missing fields rather than rejecting a rule. None of that travels with the data, and
  all of it changes what a consumer sees.

  Two honesty notes are in `evidence_limits` rather than in prose. Seven signatures are not
  malware coverage — a clean `SIG` result is not "no malware", and the seed runtime's own
  header calls the table "deliberately tight". And three of the seven match on **names**
  (`xmrig`, `mimikatz`, `meterpreter`), so a document *discussing* those tools matches; the
  seed runtime papers over this by excluding `knowledge/`, `tests/`, `docs/` and
  `node_modules/` from the scan, which is engine behaviour and does not come with the table.

  Verified: 7/7 rule objects field-identical to source including key order, no non-ASCII
  bytes, and all seven compile in Node bare, `i` and `iu` (21/21) and in Python `re` (7/7).
  Note the exact family spellings — `reverse_shell`, not `reverse-shell`, and `cryptominer`,
  not `miner`; they are policy keys, and the working note that seeded this file had both wrong.

### Verification

Every file above except `calibration.json` was proven rather than transcribed: the data was
rebuilt **from the commons JSON alone** and diffed against the source implementation. Each
file records its own result and its own limits.

`calibration/calibration.json` carries `verified: false`. Its source arrived as a prose
summary rather than as code, so no differential check was possible, and the file names the
checks that were not run instead of attaching a caveat to a pass.

The corpus was verified the same way the data was — by a harness that does **not** share the
generator's knowledge. It reads only the case directories, re-runs both runtimes on the bytes
it finds there, and checks every field of every `expected.json`, digests included: **83
cases, 0 failures**. Two further checks, because a corpus that cannot fail is not evidence:
commons' family severity matches the severity the guard emits per finding, **83/83**; and
deleting the middle third of each input breaks **76 of 83** expectations. The 7 survivors are
the shortest payloads, where the mutation leaves the trigger intact — that is a weak
mutation, not a weak fixture, and it is recorded as such rather than rounded up.

- `docs/lexicon-port-divergence.md` — informative. A differential comparison of the two
  ports of `injection-patterns.mjs` (this repository's and the Python guard's): 83/83
  patterns correspond, 64 are byte-identical, 6 differ only by escaping and are proven
  equivalent, and **13 behave differently**, with a witness input for each and misses on
  both sides. The cause is two different ReDoS mitigations of one table. **No data file was
  changed** — behaviour preservation holds and the finding is reported to the owning
  repositories.

  **Revised 2026-08-09 with one retraction.** The document claimed that *neither runtime
  misses an attack*, on the grounds that every witness payload still produced a finding. It
  does miss. That measurement ran the payloads against the **union of every pattern table
  this repository holds**, and the rescuing hit came from `active-content.json` — the Python
  guard's table. `llm-security` has no active-content table at all, so a union of commons
  tables was read as a statement about each runtime separately. Re-measured through
  `llm-security`'s own `scanForInjection()` at `b0de0ca`, all three witness payloads return
  **`found: false`** — no finding whatsoever — while controls in the same run behave
  normally. Three confirmed recall holes, which `llm-security` attributes to its v7.8.3 #24
  ReDoS hardening and has logged as a v8.x task.

  Also corrected: one of the 13 divergences does not reach report level, because the guard's
  `hybrid-xss:javascript-uri` fires on the same witness at the same severity and anchor. The
  report-level number is **12**. And the `hybrid` severity question that the document reported
  rather than resolved is now closed — the reported hint was right, the citation behind it was
  not.

  **Revised again 2026-08-10.** The document said 13 was the number blocking `conformance/`,
  since a fixture names labels. It blocks a fixture written over a **witness** input, and the
  corpus contains none — all 13 agree on their own case input. The divergence itself stands
  unresolved and unchanged; what was wrong was the claim about what it blocked.

  Corrections are marked in place rather than edited away.

### Changed

- `schema/finding.schema.json` **0.1.0 → 0.2.0** — the schema is **closed**. It was seeded
  from `sarif-formatter.mjs`, which *consumes* findings, so its property list could only ever
  be a lower bound and `additionalProperties` had to stay open. The producer is now known —
  `finding()` in `scanners/lib/output.mjs`, line 32 — and it returns an object literal with
  **exactly ten keys and no spread**: `id`, `scanner`, `severity`, `title`, `description`,
  `file`, `line`, `evidence`, `owasp`, `recommendation`. `additionalProperties` is `false`,
  and the two keys the old schema never knew about (`id`, `evidence`) are added.

  `id` gets its own definition: `DS-<prefix>-<counter>`, pattern `^DS-[A-Za-z]+-[0-9]{3,}$`.
  The `{3,}` is deliberate — `padStart(3, '0')` is a minimum, so a run emitting more than 999
  findings produces four digits. The id comes from a process-global counter, so it is stable
  neither across runs nor across processes, and the definition says so before someone keys on it.

  Nullability is now evidence rather than convention. Five keys are emitted as `null` rather
  than omitted (`opts.x || null`), so a serialised finding always carries all ten. The
  exception is the four assigned straight from `opts`: omit `description` and the key is
  `undefined` and vanishes from the JSON. Verified by calling the real producer — ten keys in
  memory, nine after serialisation.

  **`owasp` is a string, not an array, and not one code.** Multiple codes are joined with
  `, `. Measured across the seed runtime: 31 distinct values over 157 emission sites, 13 of
  them multi-code, and **four mix taxonomies inside a single value** (`LLM06, ASI02` and
  friends) with no discriminator saying which is which. That sharpens the edition problem
  `mapping/owasp-map.json` already records, and it has a consequence nobody had written down:
  `sarif-formatter.mjs` builds `tags: [f.owasp]`, so a finding anchored to two taxonomies
  produces **one** SARIF tag with a comma in it. Nothing filtering on `LLM06` will match.
  Reproduced end to end through the real `finding()` and `toSARIF()`, and logged as
  `known_lossiness.owasp-tag-not-split` — consumer behaviour in `llm-security`, not data, so
  it is reported rather than fixed here.

  The **JSONL profile is `not applicable`, not `unspecified`** — the distinction is the point.
  `unspecified` would claim a profile exists and merely has not been written down. No
  finding-JSONL exists: findings are emitted only inside a single JSON envelope
  (`output.mjs:140`). The one module that does write JSONL, `audit-trail.mjs`, writes *audit
  events* under a different schema — where `owasp` is an **array**. Same field name, different
  type, same repository. A consumer reading both through one code path will be wrong about one
  of them, so the profile records the trap instead of leaving a TODO.

  Verified: the schema is valid Draft 2020-12, every finding built by the real producer
  validates against it, and four negative controls (extra property, missing `id`, malformed
  `id`, unknown severity) are all rejected.

  One new open question, unpatched by design: the producer's JSDoc lists **seventeen** scanner
  prefixes including `IDE`, while all four maps in `mapping/owasp-map.json` are keyed on
  **sixteen** without it. An `IDE` finding has no taxonomy mapping in any map. Adding the key
  would be inventing detection data.

- `lexicon/injection-lexicon.json` **0.4.0 → 0.5.0** — the last null in the file is filled and
  the id space is ratified. Two blockers close, no detection data moves.

  `families[hybrid].severity` was `null`, deliberately, because the seed dump did not supply
  it. It is **`high`** — and the interesting part is where that is written. The hybrid family
  has no severity field anywhere; the engine assigns one by pushing `HYBRID_PATTERNS` matches
  straight into the `high` bucket at `injection-patterns.mjs:274-281`. Both this repository
  and the Python guard had first looked in `severity.mjs`, which contains no injection-family
  severity at all. The guard's port holds the right value behind that wrong citation, so
  `severity_provenance.not_from` records the miss explicitly: a wrong citation to a right
  value is the harder defect to catch later.

  > **Correction 2026-08-10 (see Unreleased):** the two sentences about *the guard's* citation
  > are false and were never measured here. The guard's port cites `injection-patterns.mjs`,
  > the right file. Everything above about `severity.mjs` containing no injection-family
  > severity, and about where the value actually lives, stands and has been re-measured.

  `pattern_id_space.not_yet_confirmed` is replaced by `ratification`. Both seeding runtimes
  agreed on 2026-08-09 — `llm-security` ratified the 0.2.0 proposal as-is and treats an id
  change as breaking on the same terms, and the guard confirmed the space its own port
  supplied. `id` is now a cross-runtime contract, which is what `conformance/` was waiting
  on to be able to name a finding.

  `alias_evidence.llm_security` is sharpened rather than upgraded. All 83 alias strings were
  confirmed equal to the module's `label` field, in order — so the alias is certainly the
  pattern's name **in the table**. It is still not established that a finding carries it: the
  producer is `output.mjs:finding()`, which emits `title` and has no `label` key at all.
  Verified at table level, one level short of where it would matter. Match on `id`.

- `lexicon/injection-lexicon.json` **0.3.0 → 0.4.0** — verified against the source module
  instead of against the dump it was transcribed from, and **two false provenance claims
  retracted**. The source is now pinned: `b0de0ca` on the public remote, imported in Node
  and compared entry by entry on `source`, `flags` and `label`.

  The result is **83/83 byte-identical to source**, which is not what the file previously
  claimed. It said two patterns had been rewritten from raw code points into `\uXXXX`
  escapes; the module already writes them escaped, so nothing had been rewritten. The stored
  pattern text was right the whole time — only the account of where it came from was wrong.
  The dump had rendered the module's escapes as the characters they denote, and this
  repository re-escaped them, arriving at the correct bytes by way of an incorrect story.

  The same inversion ran the other way in `multi-lang:french`, which carried the class
  spelled with a raw accented Latin `e` where the module writes it as the escape
  `\u00e9` inside the same character class.
  That was the one pattern of 83 not byte-identical to source,
  and it is corrected. The two spellings are the same regular expression — verified in Node
  bare and under `u`, and in Python `re`, over accented, unaccented, uppercase and
  non-matching French input, with identical match offsets — so **no behaviour moved**. No
  pattern in the file contains a non-ASCII byte now, matching the module, whose regex
  literals are pure ASCII throughout.

  Structurally: `normalisations` is now `[]` with a `normalisations_note`, matching the
  convention already used in `signatures/secret-egress.json`, and a new `source_fidelity`
  block carries the counts, the method, the verified class membership, and both retractions
  in full. Retracted claims are recorded rather than deleted — the earlier equivalence
  evidence (692 Node comparisons, 236 Python) remains true, it is simply no longer
  load-bearing.

- `lexicon/injection-lexicon.json` **0.2.0 → 0.3.0** — the two aliases are no longer presented
  as equally backed. `pattern_id_space.alias_evidence` now records each one separately:
  `llm_ingestion_guard` is **verified** (the guard's coverage matrix asserts on that exact
  string, so it is demonstrably what a guard finding carries), while `llm_security` is
  **not** — it is the pattern table's own name, and the finding producer was never supplied,
  with the known Node finding shape using `title` rather than `label`. Averaging the two into
  one file-level claim would have repeated the defect this repository corrects per-table
  elsewhere.

  Also: `normalisations[].affects` now keys on `id` with the prose names kept beside it as
  `affects_labels`. An internal cross-reference on label was a second identity space inside
  the file the id was added to unify.

- `lexicon/injection-lexicon.json` **0.1.0 → 0.2.0** — every pattern gains a commons-owned
  `id` and an `aliases` object naming what each seeding runtime calls it, plus a top-level
  `pattern_id_space` block explaining the field. This exists because a `conformance/`
  fixture has to name a finding and the two runtimes do not name the same pattern the same
  way.

  The id was **adopted verbatim from the guard's port**, which already carried both names,
  rather than invented here. Matching was by `label` ↔ `desc` with em-dash normalised to
  hyphen: 83/83, one-to-one, ids unique.

  **No detection data moved.** Labels, patterns and flags are byte-identical in sequence,
  no `flags` key was invented (78 before, 78 after), and stripping the three new fields
  reproduces the previous committed file byte for byte — 23 566 bytes, identical. All 83
  patterns still compile in Node bare and under `u` (166/166) and in Python `re` (83/83).

  Neither `llm-security` nor the guard has ratified this id space yet; both were asked by
  coord on 2026-08-09, and the file says so rather than implying agreement.

### Not included

- `spec/decode-pipeline.md` — needs the decode implementation. A normative spec inferred
  from a data dump would be worse than an absent one.
- **Conformance for the other four tables.** The corpus covers the injection lexicon only.
  The carrier, active-content and secret-egress tables have 11 convertible cases waiting in
  the guard's matrix, and no ratified cross-runtime finding id between them — writing those
  fixtures would mint a contract unilaterally, in the same stroke as the tag. Named in
  `conformance/manifest.json` under `scope_planned`.
- The 29 non-convertible cases of the guard's 134 assert a runtime's **API surface** — that
  a Python call raises `OKFPathError`, that a disposition engine composes two findings a
  particular way. This repository does not own an API, so those belong to the guard's suite.

`spec/decode-pipeline.md` is named in the README as planned rather than linked, so nothing
in the repository points at a file that does not exist.