Stats
Actions
Tags
From lumin-repo-lens
Maintainer-only surface for Lumin Repo Lens: drafts and checks canonical repository facts, detects canonical drift, and validates parser contracts with machine-evidence requirements.
How this skill is triggered — by the user, by Claude, or both
Slash command
/lumin-repo-lens:lumin-repo-lens-canonThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
This is the maintainer-facing canonical fact surface for lumin-repo-lens.
This is the maintainer-facing canonical fact surface for lumin-repo-lens.
It owns canon-draft and check-canon together because draft output is
reviewed, promoted, and later checked for drift as one lifecycle.
Default voice is concise and factual. This surface may use colder
terminology than the audit surface, but every claim still needs machine
evidence and scan range.
Do not use casual hedging. If evidence is missing or degraded, emit
unknown or a degraded status directly with the scan range.
NO STRUCTURAL CLAIM WITHOUT MACHINE EVIDENCE
NO ABSENCE CLAIM WITHOUT STATED SCAN RANGE
The shared engine lives in the sibling audit skill:
node ${CLAUDE_PLUGIN_ROOT}/skills/lumin-repo-lens/scripts/audit-repo.mjs
In a maintainer checkout, the equivalent command is:
node audit-repo.mjs
Slash commands still read <SKILL_ROOT>/references/command-routing.md from the
shared audit skill for exact flag routing.
Below, <SKILL_ROOT> means
${CLAUDE_PLUGIN_ROOT}/skills/lumin-repo-lens in plugin
installs, or the repo root in a maintainer checkout.
Below, <audit-repo> means whichever of the two commands above applies
to the current context.
This surface owns /lumin-repo-lens:canon-draft and
/lumin-repo-lens:check-canon.
<SKILL_ROOT>/references/command-routing.md first for
slash-command routing.<SKILL_ROOT>/references/lifecycle-modes.md when exact flags, exit codes, or
artifact names matter.<SKILL_ROOT>/canonical/canon-drift.md for drift categories, parser contract,
per-source Markdown reports, and canon-drift.json shape.<SKILL_ROOT>/canonical/fact-model.md,
<SKILL_ROOT>/canonical/identity-and-alias.md, and
<SKILL_ROOT>/canonical/classification-gates.md when a drift category, identity, or
label set is ambiguous.Use when current artifacts should propose canonical facts for review:
<audit-repo> --canon-draft --root <repo> --output <dir> --sources <sources>
Drafts are proposals, not promoted truth. The model may summarize why a draft changed, but a human or maintainer process still promotes it.
Use when promoted canonical/ facts should be compared against fresh
artifacts:
<audit-repo> --check-canon --root <repo> --output <dir> --sources <sources>
If a source is missing, skipped, degraded, or diagnostic-only, surface that status. Do not turn "skipped" into "clean."
Use short maintainer blocks:
For full details, cite canon-drift.json, per-source
canon-drift.<source>.md, and manifest.json.
If the user shifts to general repo structure, cleanup priority, or
refactor-plan coaching, hand off to lumin-repo-lens. If the
user shifts to code changes before or after implementation, hand off to lumin-repo-lens-write-gate.
npx claudepluginhub annyeong844/lumin-repo-lens --plugin lumin-repo-lensAudits a repo for AI-readiness, scoring ~20 dimensions across Foundation, Why, What, Hygiene, and Sync. Use when inheriting a legacy repo or asking "is this repo agent-ready?"
Audits a repository to map its real stack, conventions, assets, tests, docs, risks, and integration points. Persists results in reusable markdown to reduce re-reading and save tokens. Also calculates a harnessability score (0-100) to assess how well the codebase supports autonomous agent work.
Audits a repository for baseline compliance across 9 categories including code quality, security, CI/CD, testing, and documentation. Emits Markdown report and JSON sidecar.