D12 (org-ops docs/decisions.md, 2026-08-01) requires the plugin-class
disclosure line to be self-contained with three parts: generator, process,
and the ownership basis. Voyage carried the first two; adds the Anthropic
Consumer Terms §4 clause so the line no longer depends on a cross-repo
anchor that does not exist.
repo-standard v0.3.0: 14/14 OK. Suite: 832 (830 pass / 0 fail / 2 skipped).
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01WfGRt8MkPVvnkGCzidTuis
The `[Full disclosure →](../../README.md#ai-generated-code-disclosure)`
link was inherited from the monorepo layout. In a standalone repo `../../`
resolves outside the repository (404), and the anchor never existed in any
README. Replaced with the inline sentence the polyrepo migration intended.
Reported by catalog (coord 20260803T191159Z-4541213484), confirmed by
repo-standard v0.2.0 gate (LINK-OUTSIDE-REPO, README.md:11).
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01PRYUCVByhmzWBMCCFSYK2j
repo-standard gate (plugin class) flagged 6 ERROR. Fixed all six:
- README-DESC: H1 is now `# voyage` (was `# trekplan — …`, stale from the
rename) and the opening line is byte-identical to the forge description,
so description == catalog == README on all three surfaces.
- INSTALL-NO-CLI + HEADING-MISSING: new first-screen `## Install` with the
CLI install command (`claude plugin install voyage@ktg-plugin-marketplace`)
beside `marketplace add`; the `enabledPlugins` JSON is kept as the second
form. Retires the duplicate `## Installation` section further down (its
architect paragraph already exists verbatim under Quick start) and drops
the now-redundant install lines from the Quick start block.
- HEADING-MISSING: `## Non-goals` (promotes the existing "What it doesn't
solve" list) and `## Changelog` (release notes moved off the first screen,
per the standard, now a link + one line per recent version — keeps the
doc-consistency pin on framing / brief_version 2.2).
Remaining 3 gate ERROR are a classifier defect in the gate, not this repo:
directory link targets (`agents/`, `examples/observability/`) are checked
against `git ls-files`, which never contains directories. Links are correct
and left as-is by operator decision; reported to repo-standard via coord.
Suite: 832 tests, 829 pass, 1 fail, 2 skipped — the failure is the
load-dependent otel tail-latency NFR (passes 3/3 in isolation), unrelated.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01PRYUCVByhmzWBMCCFSYK2j
Phase 3 (atomic-write) and Phase 4 (validator) used !`...` eager-exec
with unresolved runtime placeholders; the harness executes those at
command LOAD time, so zsh parsed <project-dir> as input redirection and
/trekendsession aborted before the model saw a single instruction.
Both blocks are now plain runtime Bash fences with {curly} placeholders
(shell-inert, trekplan.md convention) and absolute ${CLAUDE_PLUGIN_ROOT}
paths (cwd-relative plugin paths were a latent ERR_MODULE_NOT_FOUND in
any user repo). Phase 1 discovery block keeps its legitimate eager-exec.
Regression guard: new tests/commands/trekendsession.test.mjs flags any
!`-block in commands/*.md containing <angle>/{curly} placeholders, and
pins Phase 3/4 as runtime Bash. Suite 828 -> 832 (830/0/2). E2E: fixed
blocks run with real values write both state files, validator valid:true;
trekcontinue.md:147 runtime-verified self-contained.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_013NXzF3F2zAM8S7FCBXCqAb
Operator reviewed the brief (S57) and approved as-is, then chose research-first
at the /trekplan research gate (option A: investigate only the genuinely-external
topic 1; fold local topics 2/3 into /trekplan exploration).
Topic 1 finding (validator-green; claude-code-guide + CC 2.1.196 binary + a real
local /deep-research run, all cited): /deep-research is a built-in *dynamic
workflow* (not a skill), outside the Skill-tool allowlist. Trigging is
prose-instruction only; the report lands inline in context with no on-disk
artifact (only the .js script is written). => the engine must be instruction-based
+ in-context transform, surface-only. Brief's v2.1.154+ / Pro-via-/config
constraints verified correct; SC3 ("dynamic workflows off -> fallback to swarm")
is correct as written (an earlier review note that called it a conflation rested
on a wrong premise and is retracted).
- docs/deep-research-engine-research.md: new, research-validator green strict.
- docs/deep-research-engine-brief.md: status draft->ready (operator-approved),
research_status pending->complete (option-A decision recorded), Research Plan
traceability note.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_019ddpUq3MNQFoV5h3qYrVYj
Operator-delivered draft (S55) carried only brief_version/status/brief_quality/
research_topics and an explicit "reconcile frontmatter against brief-validator
before /trekplan" note. Add the required fields (type, task, slug,
research_status, phase_signals_partial) plus the brief_version 2.2 gates:
framing: refine (operator-confirmed — additive opt-in engine, swarm stays
default, no contract change) and a 5-line ## TL;DR. brief-validator passes
strict + --min-version 2.2. status stays draft pending operator review before
the /trekplan run.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01WH1krHamUehZh6JqVqs85t
The documented `brief-validator.mjs <brief.md>` invocation (no flags) always
bailed to Usage/exit 2. Root cause: when --min-version is absent, minIdx is -1,
so the skip index minIdx+1 was 0 — excluding argv index 0, exactly where the
file positional sits in the no-flag case. Any leading flag (--soft, --json)
pushed the file to index >=1 and masked the bug, so the function-level tests
never caught it.
Guard the skip index to -1 when --min-version is absent. Add two CLI regression
tests (execFileSync, matching the next-session-prompt-validator pattern):
no-flag invocation reaches validation, and --min-version still skips its value
token to find the file. Suite 822 -> 824 (822/0/2).
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01WH1krHamUehZh6JqVqs85t
Version sync 5.7.1 -> 5.8.0 across plugin.json, package.json,
package-lock.json, README badge, and CHANGELOG top entry (guarded by
doc-consistency.test.mjs). Suite 822 (820/0/2).
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01BJQYC5vpkJWxndS55vQQZ6
Records the operator-supplied brief for an --engine {swarm|deep-research}
choice on /trekresearch's external phase (swarm stays default; deep-research
delegates the external phase to the built-in workflow with auto-fallback).
Backlog item only — not implemented. Lives beside the work (docs/), tracked.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01BJQYC5vpkJWxndS55vQQZ6
Scores committed agent-run fixtures against the golden corpus at
(file, rule_key) granularity, building on the deterministic coordinator
contract (4a). Offline: committed reviewer payloads, no live agent spawn,
no LLM, no network (the LLM-in-the-loop grading is the separate 4c tier).
- lib/review/gold-scorer.mjs: scoreFindings (precision/recall/f1 at
(file,rule_key) granularity, line+severity ignored) + scoreVerdict; pure,
with documented vacuous-set conventions.
- tests/fixtures/bakeoff-rich/runs/run-perfect.json: committed run that
reproduces all 5 seeded gold findings through runContract.
- tests/lib/gold-eval.test.mjs: the scoring RUN (precision/recall/f1 = 1.0,
verdict == expected_verdict BLOCK, nothing suppressed/skipped).
- lib/util/test-census.mjs: third census category (goldEval) — a scoring run
is neither behavior coverage nor a doc-pin; honest-count invariant now 3-way.
- docs/eval-corpus/README.md: 4b moved from Future hardening to implemented.
Suite 809 -> 822 (820/0/2). gold-scorer covers TP+FP+FN+degenerate paths.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01BJQYC5vpkJWxndS55vQQZ6
SC1 (zero `ultra` refs) was a pre-existing false-positive on
docs/cc-upgrade-2.1.181-decision-matrix.md, which cites real CC features:
`ultracode` (a CC keyword, 2.1.160) and the `ultra-cc-architect` plugin
name. Rewording would make the doc factually wrong, so the file is
excluded via exclude_path() — same pattern as CHANGELOG/MIGRATION.
Tooling-only (verify.sh); no version bump. verify.sh 7/0, suite 809 (807/0/2).
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_013J12WFAbngQNMTJMoybD7N
origin is open/voyage (an OFFENTLIG/public mirror per ~/.claude/CLAUDE.md, which
names open/ as the public example). STATE.md was wrongly tracked since 5bf574a
(23 commits) under a mistaken 'PRIVAT Forgejo' premise. Its content has now been
scrubbed from all history via git filter-repo; this gitignores it going forward
and corrects the false tracked/private comments. STATE continuity stays local.
Cross-session coordination brief from the config-audit machine-tuning session,
dropped into voyage per operator instruction. brief_version 2.2, framing: refine.
Locked as the next session's task (Alternative A). Tracking it makes the next
session's /trekplan --brief input deterministic and durable on the private remote.
F1 cache-analyzer regression guard (SC5): pin percentile (wall_time_ms_p50/p90)
and time-range (oldest/newest_event_iso) — 2 of 3 'unchanged' categories were
previously un-asserted on the mixed-input fixture.
F2 lastMainChainModel: add direct coverage — last-wins across 2 distinct
main-chain models, sidechain exclusion (even when the sidechain is the last
record), and model-absent → null propagating to deriveCost refuse-to-estimate.
804 -> 807 tests (805 pass / 0 fail / 2 skipped). No test pins the test count.
