Commit graph

15 commits

Author SHA1 Message Date
96ac93cc3a feat(voyage): add 5 additive profile fields to JSONL stats — SC #11
Step 8 av v4.1-execute (Wave 3, Session 4).

5 nye additive felter er nå dokumentert i hver kommandos prose-stats-blokk
(via Profile-seksjonen fra Step 7 — felles overflate per kommando):
- profile           — string ('economy' | 'balanced' | 'premium' | <custom>)
- phase_models      — object form {brief: 'sonnet', ..., continue: 'opus'}
- parallel_agents   — number (snapshot av maksverdi som faktisk ble brukt)
- external_research_enabled — boolean
- profile_source    — 'flag' | 'env' | 'default' | 'inheritance'

Patcher trekresearch.md med eksplisitt profile_source-mention + alle 5 felter
(de andre 5 commands hadde dette allerede via Step 7 Profile-seksjon).

SC #11 contract-test design (per brief):
(a) Fixture-records valideres som JSONL-contracts → tests/fixtures/stats-with-profile.jsonl
    (5 simulerte stats-rader, én per kommando-overflate)
(b) Command-prose contains field-names → kompenserer for plan-critic Major 4
    false-confidence (faktisk runtime-emission er LLM-prose-driven, ikke
    testbart i node:test alene).

Tester (12 nye, baseline 445 → 457):
- Fixture parses som JSONL (5 records)
- Hver record har profile + profile_source
- profile_source-verdier i {flag, env, default, inheritance}
- Fikstur dekker alle 4 profile_source-verdier
- 6 commands × prose contains profile + profile_source
- trekplan.md prose contains phase_models + parallel_agents
- trekresearch.md prose contains external_research_enabled

Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
2026-05-09 09:40:21 +02:00
4e6b026104 feat(voyage): add lib/exporters/textfile-format.mjs — Prometheus text-format pure transform SC #12
Step 9 av v4.1-execute (Wave 2, Session 3).

Pure function transformToPrometheus(records) → Prometheus text-format 0.0.4.

Hard rules:
- NO client-side timestamps (research/01 node_exporter#1284 mitigation)
- Allowlist-redacted records ONLY (caller responsibility — Step 11 enforces)
- UTF-8 metric names normalized: lowercase, [.\\-\\s] → _, voyage_ prefix
- Empty input → empty string output
- Sorted output for determinism (snapshot-test-friendly)

Heuristic metric typing:
- counter: *_total, *_count, *_passed, *_failed, *_skipped
- histogram: *_ms, *_duration, *_p\\d+, *_seconds
- gauge: everything else (Prometheus convention)

Snapshot: tests/fixtures/expected.prom byte-for-byte match.
Regenerate: node scripts/gen-expected-prom.mjs > tests/fixtures/expected.prom

Tester (6 nye, baseline 400 → 406):
- Snapshot byte-for-byte match (SC #12)
- Empty input handling (null, undefined, [])
- Allowlist-redaction sanity (post-bash-stats uten command_excerpt)
- NO client-side timestamps (token-count-assertion per linje)
- normalizeMetricName edge-cases
- Determinism (identisk input → identisk output)

Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
2026-05-09 09:30:58 +02:00
f8fc0bb1c2 feat(voyage): add lib/profiles/resolver.mjs — locked interface SC #5-#9
Step 6 av v4.1-execute (Wave 2, Session 2).

Implementer locked interface contract fra brief Preferences:

- loadProfile(name, opts) → ProfileObject
  Leser lib/profiles/<name>.yaml (built-in) eller custom fra
  <cwd>/voyage-profiles/ > ~/.claude/voyage-profiles/. Throws Error med
  cause: PROFILE_NOT_FOUND. Returnerer parsed object med phase_models
  flattened til {brief: 'sonnet', research: 'opus', ...} (object form
  for downstream JSON-stats).

- resolveProfile(argv, env) → {profile, profile_source}
  Ordre: --profile flag > VOYAGE_PROFILE env > 'premium' default.

- resolveTrekcontinueProfile(planPath, argv, opts) → {profile, profile_source}
  --profile flag wins ('flag'); ellers leser plan.md frontmatter
  ('inheritance'); v4.0-stil plan uten profile-felt → 'default' premium
  (backward-compat). Flag overstyrer arv → console.error advisory.

- validateProfileFile(path) → Result
  Tynn re-eksport av validateProfile fra profile-validator.mjs.

- findProfilePath(name, opts) → {path, attempted}
  Lookup-helper. attempted-array brukes i error-melding for HIGH-risk-
  mitigering (ENOENT-diagnose).

Tester (13 nye, baseline 387 → 400):
- SC #5 x4 (loadProfile economy/balanced/premium + PROFILE_NOT_FOUND)
- SC #6 (flag > env > default ordre)
- SC #7 (performance: 1000-iter < 50ms gjennomsnitt; faktisk ~0.055ms)
- SC #8 x2 (cwd > home precedence + error-msg attempted-paths)
- SC #9 x2 (inheritance + flag-override-advisory)
- Backward-compat x2 (v4.0 plan + non-existent plan)
- validateProfileFile re-export sanity

Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
2026-05-09 09:29:01 +02:00
b56ba13b04 feat(voyage): add lib/validators/profile-validator.mjs — SC #1, #2, #3
Step 5 av v4.1-execute (Wave 2, Session 2).

Profile-validator etter brief-validator-mønsteret eksakt: validateProfileContent
(pure), validateProfile (file-reader), CLI shim med --json flag. Eksporter
PROFILE_REQUIRED_FIELDS (frozen), PROFILE_REQUIRED_PHASES (frozen).

Validerer:
- Required frontmatter fields (name, phase_models, parallel_agents_min/max,
  external_research_enabled, brief_reviewer_iter_cap)
- phase_models = list-of-dicts med phase + model
- 6 required phases (brief, research, plan, execute, review, continue)
- parallel_agents_max ≥ parallel_agents_min
- Allowed model values: ['sonnet', 'opus']; haiku tillatt KUN ved
  VOYAGE_ALLOW_HAIKU=1 (per global CLAUDE.md modellvalg-prinsipp)

Issue codes: PROFILE_MISSING_FIELD, PROFILE_INVALID_MODEL, PROFILE_INVALID_ENUM,
PROFILE_READ_ERROR, PROFILE_NOT_FOUND.

Field-path-reporting i error-location: phase_models[N].model for SC #2.

Tester (10 nye, baseline 377 → 387):
- SC #1 x3 (innebygde profiler grønne)
- SC #2 (PROFILE_INVALID_MODEL med location phase_models[2].model)
- SC #3 (PROFILE_INVALID_ENUM for external_research_enabled: "yes" string)
- VOYAGE_ALLOW_HAIKU env-var deny/allow
- PROFILE_MISSING_FIELD når name fraværende
- PROFILE_NOT_FOUND for ikke-eksisterende fil
- 2 export drift-pins

Fixturer: profile-invalid-model.yaml (gpt-4 i phase_models[2]),
profile-invalid-enum.yaml (external_research_enabled som string "yes").

Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
2026-05-09 09:26:23 +02:00
43a3e4f11d docs(voyage): jsonl schema audit — field-allowlist input for v4.1 otel exporter
Step 1 av v4.1-execute (Wave 1, Session 1).

Audit alle 6 trek*-stats.jsonl-skjemaer + lib/stats/event-emit.mjs autonomy
events + hooks/scripts/post-bash-stats.mjs PostToolUse Bash records. Produser
markdown-tabell {schema_id, fields[], writer_path, line_ref, v4.1 additive,
PII} som load-bearing input til Step 11 (field-allowlist) og Step 8 (stats
plumbing).

Spesielle merker:
- command_excerpt fra post-bash-stats.mjs flagget CWE-212 (improper cross-
  boundary removal of sensitive data) — eksporten MÅ hard-ekskludere uten
  eksplisitt VOYAGE_EXPORT_INCLUDE_COMMAND_EXCERPT=1 (deferred til v4.2)
- v4.1 additive fields enumerert per skjema: profile, phase_models,
  parallel_agents, external_research_enabled, profile_source
- EXPORT_ALLOWLIST + EXPORT_DENYLIST utdrag i bunnen som forhåndsdefinisjon
  av Step 11 inline static consts

Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
2026-05-09 09:20:54 +02:00
c909466c99 feat(voyage)!: bulk content rewrite ultra -> voyage/trek prose [skip-docs]
Sed-pipeline (16 patterns, longest-match-first) sweeper residuelle ultra*-treff
i prose, command-narrativ, agent-prompts, hook-kommentarer, doc-prosa.

Pipeline-utvidelser fra V4-prompten:
- BSD-syntax [[:<:]]ultra[[:>:]] istedenfor \bultra\b (BSD sed mangler \b)
- 6 compound-patterns for ultraplan/ultraexecute/ultraresearch/ultrabrief/
  ultrareview/ultracontinue uten -local-suffiks
- ultra*-stats glob -> trek*-stats glob
- Linje-eksklusjon redusert til ultra-cc-architect (Q8); session-state-
  eksklusjonen var over-protektiv
- File-eksklusjon utvidet til settings.json, package.json, plugin.json,
  hele .claude/-treet (gitignored + V5-territorium)

Q8-undantak holdt: architecture-discovery.mjs + project-discovery.mjs urort.
Filnavn-konvensjon holdt: .session-state.local.json + *.local.* preservert.

Manuell narrative-fix: tests/lib/agent-frontmatter.test.mjs linje 10
mangled "/ultra*-local" til "/voyage*-local" (ingen slik kommando finnes);
korrigert til "/trek*".

Residualer utenfor scope (V5 handterer): package.json + .claude-plugin/
plugin.json (Step 12-14 versjons-bump). .claude/* er gitignored
spec-historikk med tilsiktet BEFORE/AFTER-narrativ.

Part of voyage-rebrand session 3 (Wave 4 / Step 10).

Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
2026-05-05 15:08:20 +02:00
12f4391dca feat(voyage)!: rename type discriminators across validators + fixtures [skip-docs]
- brief-validator: BRIEF_TYPE_VALUES ['ultrabrief','ultrareview'] -> ['trekbrief','trekreview'] + dependent branches
- research-validator: 'ultraresearch-brief' -> 'trekresearch-brief'
- review-validator: 'ultrareview' -> 'trekreview'
- 3 templates frontmatter type:
- 4 synthetic fixtures: ultraplan-synthetic/ultrareview-synthetic -> trek* (frontmatter only; bodies untouched, Jaccard floor preserved)
- 2 trekreview fixtures: type: trekreview
- 6 validator-test fixtures + asserts
- agents/review-coordinator.md frontmatter example

Atomic: validator + fixtures committed together — partial state would cause vacuous
test passes or hard validator rejection.

Part of voyage-rebrand session 2 (W3.3 / Step 5).

Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
2026-05-05 14:40:25 +02:00
4f8d9c70e5 feat(voyage)!: rename commands, templates, fixtures for v4.0.0 [skip-docs] 2026-05-05 14:13:44 +02:00
Kjell Tore Guttormsen
9a285208d2 test(ultraplan-local): add session-state fixtures
Two fixtures for session-state-validator (Step 4):
- valid-in-progress.json — well-formed schema-v1 object
- malformed.json — truncated JSON for negative tests

Step 3 of /ultracontinue v3.3.0.
2026-05-01 20:21:50 +02:00
Kjell Tore Guttormsen
600bde6a41 test(ultraplan-local): add SC3(b) source_findings structural test
Synthetic plan.md fixture with source_findings: block-style YAML list of 3
40-char hex IDs in frontmatter, plus minimal plan structure (Title +
Implementation Plan + 1 Step + Manifest). 3 tests verify:

1. plan-validator accepts a plan with source_findings (additive optional field)
2. frontmatter parser extracts source_findings as array of strings
3. each ID matches the 40-char lowercase hex format from finding-id.mjs

Closes the SC3(b) gap flagged by adversarial review (scope-guardian Gap 2).
LLM-level behavior (planner emitting source_findings) remains non-testable
without live invocation; this covers the structural contract.

Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
2026-05-01 17:23:30 +02:00
Kjell Tore Guttormsen
68a0e5aa3b test(ultraplan-local): add synthetic ultrareview determinism fixtures
review-run-A.md (5 findings) and review-run-B.md (6 findings, A ⊂ B) form a
known-Jaccard fixture pair: |A ∩ B| = 5, |A ∪ B| = 6, Jaccard = 5/6 = 0.833,
above the SC4 threshold of 0.70. IDs are real 40-char SHA1s computed via
lib/parsers/finding-id.mjs from realistic (file, line, rule_key) triplets.

Both fixtures pass review-validator --strict (frontmatter + body sections +
findings shape). Real-LLM determinism measurement deferred to v1.1.

Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
2026-05-01 17:21:02 +02:00
Kjell Tore Guttormsen
b787213ba8 chore(release): ultraplan-local v3.1.0
Quality program release. Spor 0+1+2+3 all delivered.

- 109 zero-dep tests gate fork-readiness
- 5 validators wired into 4 commands as CLI shims
- HANDOVER-CONTRACTS.md: single source of truth for 5 pipeline handovers
- PreCompact-hook (P0) closes progress.json drift; --resume now works
- Semantic plan-critic catches paraphrased deferred decisions
- examples/01-add-verbose-flag/: hand-calibrated end-to-end pipeline demo
- 4 hooks total (pre-bash, pre-write, session-title, post-bash-stats, pre-compact-flush)
- SECURITY.md + Extending-the-plugin docs

CC v2.1.x feature adoption: F8 (MCP_CONNECTION_NONBLOCKING),
F9 (sessionTitle), F3 (duration_ms), F12 (disableSkillShellExecution).
F2 (hook 'if'-field) deferred — universal protection wins.

Pre-flight verification:
- npm test → 109 pass
- plan-validator --strict templates/plan-template.md → READY
- plan-validator --strict tests/fixtures/plan-fase-narrative.md → FAIL (expected)
- grep smallCodebase|mediumCodebase|largeCodebase → 0 hits

Version bumped: package.json, plugin.json, README badge, root README,
root CLAUDE.md.

Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
2026-05-01 06:31:42 +02:00
Kjell Tore Guttormsen
3c2bda87db refactor(marketplace): split cc-architect from ultraplan-local into its own plugin
Extract `/ultra-cc-architect-local` and `/ultra-skill-author-local` plus all 7
supporting agents, the `cc-architect-catalog` skill (13 files), the
`ngram-overlap.mjs` IP-hygiene script, and the skill-factory test fixtures
from `ultraplan-local` v2.4.0 into a new `ultra-cc-architect` plugin v0.1.0.

Why: ultraplan-local had drifted into containing two distinct domains — a
universal planning pipeline (brief → research → plan → execute) and a
Claude-Code-specific architecture phase. Keeping them together forced users
to inherit an unfinished CC-feature catalog (~11 seeds) when they only
wanted the planning pipeline, and locked the catalog and the pipeline into
the same release cadence. The architect was already optional and decoupled
at the code level — only one filesystem touchpoint remained
(auto-discovery of `architecture/overview.md`), which already handles
absence gracefully.

Plugin manifests:
- ultraplan-local: 2.4.0 → 3.0.0 (description + keywords updated)
- ultra-cc-architect: new at 0.1.0 (pre-release; catalog is thin, Fase 2/3
  of skill-factory unbuilt, decision-layer empty, fallback list still
  needed)

What stays in ultraplan-local: brief/research/plan/execute commands, all
19 planning agents, security hooks, plan auto-discovery of
`architecture/overview.md` (filesystem-level contract, not code-level).

What moved (28 files via git mv, R100 — full history preserved):
- 2 commands, 8 agents, 1 skill catalog (13 files), 2 scripts, 8 fixtures

Documentation updates: plugin CLAUDE.md and README.md for both plugins,
root README.md (added ultra-cc-architect section, updated ultraplan-local
section), root CLAUDE.md (added ultra-cc-architect to repo-struktur),
marketplace.json (registered ultra-cc-architect), ultraplan-local
CHANGELOG.md (v3.0.0 entry with migration guidance).

Test verification: ngram-overlap.test.mjs passes 23/23 from new location.

Memory updated: feedback_no_architect_until_v3.md now points at the new
plugin and reframes the threshold around catalog maturity rather than an
ultraplan-local milestone.

Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
2026-04-30 17:18:47 +02:00
Kjell Tore Guttormsen
d59dd84377 fix(ultraplan-local): v2.3.2 — skill-drafter slug-collision hint
skill-drafter now reads {catalog_root}/<slug>.md before writing its
draft and prepends a warning block to its confirmation output when
an existing skill would be overwritten during manual `mv` promotion.
The draft is still written to .drafts/<slug>.md — the check is a
hint, not a block.

Closes v2.3.0 dogfood finding (post_dogfood_findings[0]): the
drafter produced .drafts/hooks-pattern.md when an approved
hooks-pattern.md seed already existed, giving no signal that `mv`
during promotion would silently overwrite the seed. v2.3.1
introduced the qualified-slug mechanism to resolve such collisions;
v2.3.2 surfaces them at the right moment — before promotion.

Changes:
- agents/skill-drafter.md — new Step 2 between slug computation and
  source reading. Reads {catalog_root}/<slug>.md, inspects
  review_status, derives a kebab-case qualifier from the concept
  handle (or source basename fallback). Subsequent steps renumbered
  3→7. Output format gains Collision: field and optional warning
  block. New Hard Rule.
- tests/fixtures/skill-drafter/slug-collision-expected.md — reference
  fixture documenting expected confirmation shape across four
  scenarios (no collision, approved collision, soft pending
  collision, collision with no good qualifier). Skill-drafter is
  prompt-driven; fixture anchors shape for human verification and
  downstream parsers.
- CHANGELOG [2.3.2], plugin.json 2.3.1→2.3.2, README badge, plugin
  CLAUDE.md slug-convention Collision-hint bullet, marketplace root
  README summary, marketplace root CLAUDE.md plugin table.

Non-breaking. No frontmatter/drafts-layout/tool-scope/regex changes.
Existing pipelines see one extra field and an optional warning —
both purely additive.

Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
2026-04-18 18:06:37 +02:00
Kjell Tore Guttormsen
ce04869d12 test(ultraplan-local): add skill-factory calibration fixtures
3 source/draft pairs that pin n-gram-overlap verdicts to representative prose:
- accepted (containment 0.014 / longestRun 3)
- needs-review (containment 0.211 / longestRun 12)
- rejected (containment 0.676 / longestRun 74)

Topics: session-start hooks, subagent delegation, output styles — proximate to
the production source material the skill-factory will ingest.

Plan: .claude/projects/2026-04-18-skill-factory-fase-1-mvp/plan.md (step 5)
2026-04-18 15:16:28 +02:00