refactor(discipline): rename re-anchor plugin to discipline - #1276
Conversation
Three verified defects in the old name. It was a bare verb — the sole
outlier among 60 sibling plugins, all nouns or noun-phrases. The `re-`
prefix presumes a prior anchoring, but the shared method's documented
conversation-start case is a first posture-set with nothing yet to
re-anchor. And "anchoring" is the cognitive-bias term of art, so the
plugin was named after the bias `reason-dont-recite` exists to fight.
`discipline` is the word the plugin already used for itself 189 times
("a drift corrector for research discipline") — the authors' revealed
vocabulary rather than a coinage. Container/member word overlap
(`/discipline:sweep-all`) is routine here: `/planning:plan`,
`/debugging:debug`, `/visualization:visualize`, and `/work-items:work`
all ship that shape.
Rejected on evidence: `steering` (fails the conversation-start case;
collides with Codex "mid-turn steering" and ML "activation steering"),
`calibration` (implies adjustment against a measurable reference with
quantified error — this work is judgment-based, the same objection that
rules out `invariants`), `salience` (accurate but not a word a reader
reaches for first), `grounding` (slug already taken).
Also renames `sweep-all-disciplines` to `sweep-all`, removing the one
awkward pairing the plugin rename introduced.
Migration is automatic via the marketplace `renames` map (CC >= 2.1.193);
no user action required. `CHANGELOG.md` history and `docs/topics/` are
deliberately not rewritten — they record releases that shipped under the
former name.
The verb "re-anchor" is retained in skill bodies and trigger phrases: the
defects above concern its use as a container name, not as the verb for
step 1 of the loop.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01GSXnCLnmzk8y4cKv2y1z9f
|
Claude finished @kyle-sexton's task in 2m 31s —— View job PR Review:
|
|
Claude finished @kyle-sexton's task in 1m 30s —— View job Security Review — PR #1276
|
![chatgpt-codex-connector[bot]](https://avatars.githubusercontent.com/in/1144995?s=60&v=4){kind=small}
There was a problem hiding this comment.
💡 Codex Review
Here are some automated review suggestions for this pull request.
Reviewed commit: 535d250e64
ℹ️ About Codex in GitHub
Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you
- Open a pull request for review
- Mark a draft as ready
- Comment "@codex review".
If Codex has suggestions, it will comment; otherwise it will react with 👍.
Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".
…ename The rename pass updated every `/re-anchor:*` slash-command invocation but missed the places the plugin name appears as a COMMAND ARGUMENT or a title, where the old name now resolves to nothing: - `README.md` H1 still read `# re-anchor` - `/plugin install re-anchor@melodic-software` — would fail for any new installer following the README - `/plugin configure re-anchor` in `README.md` and `setup/SKILL.md` — would fail for any user following the setup skill's own instructions - `setup/SKILL.md` prose reporting "re-anchor's effective configuration" - `setup/evals/evals.json` expected output and criteria asserting the assistant routes to the stale command `setup`'s trigger phrases gain `discipline` spellings alongside the `re-anchor` ones, which stay for users who still type the former name. The 0.9.0 CHANGELOG entry also claimed migration needed no user action. The `renames` map migrates the installed plugin record and nothing else: it cannot rewrite qualified invocations stored in a consuming repository, an agent prompt, or automation, and the `sweep-all-disciplines` -> `sweep-all` leaf rename has no compatibility path at all because the map keys plugins, not skills. The entry now states both explicitly. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01GSXnCLnmzk8y4cKv2y1z9f
|
Claude finished @kyle-sexton's task in 1m 29s —— View job PR Review:
|
|
Claude finished @kyle-sexton's task in 2m 0s —— View job Security Review — PR #1276
Result: No security issues foundThis PR is a mechanical rename — 45 files moving content from GitHub Actions —
SKILL.md instruction files — The skills that dispatch subagents (
Path references — All Cross-plugin updates — The seven modified files in sibling plugins ( Secrets and credentials — None present in any changed file. No Supply chain — No new dependencies, no new npm packages, no new GitHub Action pins or version bumps, no new remote MCP servers added. |
There was a problem hiding this comment.
💡 Codex Review
Here are some automated review suggestions for this pull request.
Reviewed commit: 930c97a470
ℹ️ About Codex in GitHub
Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you
- Open a pull request for review
- Mark a draft as ready
- Comment "@codex review".
If Codex has suggestions, it will comment; otherwise it will react with 👍.
Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".
Review response — all findings CONFIRMED, fixed in
|
| # | Finding | Verdict | Fix |
|---|---|---|---|
| 1 | plugins/discipline/README.md:1 H1 reads # re-anchor |
CONFIRMED | Now # discipline |
| 2 | setup/SKILL.md:73,81 instruct /plugin configure re-anchor |
CONFIRMED — functional; a user following the instruction hits not-found | Both now discipline |
| 3 | setup/SKILL.md:18 prose "re-anchor's effective configuration" |
CONFIRMED | Now "discipline's" |
| Codex P2 (a) | Same /plugin configure re-anchor class, plus README:363 and the eval expectations |
CONFIRMED | All updated |
| Codex P2 (b) | CHANGELOG claimed migration needs no user action | CONFIRMED — the claim was false as written | Entry rewritten |
Two more of the same class, found while verifying yours:
plugins/discipline/README.md:340—/plugin install re-anchor@melodic-software. The install instruction itself would fail for a new consumer. Higher impact than finding 1, and neither review caught it.setup/evals/evals.json:8,12—expected_outputand anexpectationsentry both asserted the assistant routes to/plugin configure re-anchor. A correctly-behaving assistant would have failed the eval.
On the setup trigger-phrase observation — agreed that keeping the re-anchor spellings is right, but keeping only them left a gap: a user typing "configure discipline" matched nothing. The description now carries both sets. The old phrases are preserved, so skill-quality:check's trigger-keyword-preservation gate against HEAD still passes.
On the retain-compatibility-commands alternative (Codex): considered, not taken. A duplicate sweep-all-disciplines skill would consume skill-listing budget permanently to serve a one-time transition. Documenting the manual step is proportionate; the CHANGELOG entry is what makes a shim discoverable if a third-party consumer ever appears.
Root cause of the class. The original sweep searched for /re-anchor: invocations and plugins/re-anchor paths. It found zero — correctly — because the misses were neither shape: the plugin name as a bare argument to /plugin install|configure, and as a document title. My own pre-push grep did hit setup/SKILL.md and I classified it as verb usage without reading the surrounding line. That is exactly how the class survived.
Verification after the fix:
claude plugin validate .— passesmarkdownlint-cli2on all changed markdown — 0 errorsevals.jsonandmarketplace.json— parse- Repo-wide search for the plugin name as a command argument (
/plugin install|configure|enable|disable,re-anchor@) — zero hits outside CHANGELOG history
…st upstream The 0.9.0 entry said the map "rewrites nothing else" beyond the installed plugin record. That is wrong. Official docs, the in-binary marketplace schema, and the decompiled migration function all agree the map rewrites the old key to the new one in the user, project, and local settings scopes for BOTH `enabledPlugins` and `pluginConfigs` — so a configured consumer's `batch_exclude` / `batch_promote` / `batch_demote` / `research_deep_verification` values move automatically. Two edges the upstream docs do not spell out, both read off the migration function: managed and policy scopes are read-only and are not rewritten, and a pre-existing `pluginConfigs` entry under the new id wins — the old entry is deleted, not merged. Below 2.1.193 the rename does not migrate at all, which reports `plugin-not-found` rather than losing configuration. The genuinely manual step is unchanged and stays documented: invocations stored outside settings are never rewritten. Source: https://code.claude.com/docs/en/plugin-marketplaces#rename-or-remove-a-plugin Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01GSXnCLnmzk8y4cKv2y1z9f
|
Claude finished @kyle-sexton's task in 1m 11s —— View job PR Review —
|
|
Claude finished @kyle-sexton's task in 1m 56s —— View job Security Review — PR #1276
Result: No security issues foundThis PR is a mechanical rename — 45 files moving content from GitHub Actions —
Path references — All Cross-plugin updates — The seven modified files in sibling plugins ( Secrets and credentials — None present in any changed file. Supply chain — No new dependencies, no new npm packages, no new GitHub Action pins or version bumps, no new remote MCP servers added. |
Rebasing onto a moved default branch invalidated two statements in the brief that were accurate when written: - #1276 renamed the `re-anchor` plugin to `discipline` and its sweep skill to `sweep-all`. The TLDR now cites the live surface, `/discipline:sweep-all`. - #1261 added a fifteenth contract slice, so the deferred question's count of persisting slices moves from fourteen to fifteen. Counted on the default branch, which is what the question is about — this branch's own slice is the sixteenth and is not yet merged. Co-authored-by: Claude <noreply@anthropic.com>
S13 cited the source article at its .work/ path, which no clone contains; it now points at the committed copy. The D-12 evidence pointer named a file that lives only in the memory slice and said nothing about that - it now states where the file is and that a clone cannot reach it, rather than dangling. Records the #1276 rename as an erratum instead of rewriting the digests. Every plugins/re-anchor path in them resolves nowhere now, but those are dated measurements taken before the rename; retargeting paths and line numbers that were never re-measured would fabricate precision. The lanes resuming against those targets are who this warning is for. Co-authored-by: Claude Opus 5 (1M context) <noreply@anthropic.com>
…sweep missed (#1365) ## Summary Fixes the two genuinely-live follow-ups from #1328 — three prose sites the `re-anchor` -> `discipline` token sweep (#1276) missed, leaving stale `re-anchor` prose beside the already-renamed token: - `plugins/work-items/reference/pipeline-shape.md:52` — "re-anchor slot" / "re-anchor set" -> "discipline slot" / "discipline set", matching the sibling `docs/conventions/loop-lane/README.md:181` the same rename commit already updated. - `plugins/work-items/skills/work-loop/SKILL.md:176` — "presence-gated re-anchor sweep" -> "presence-gated discipline sweep", matching `plugins/source-control/skills/babysit-loop/SKILL.md:319`. - `plugins/session-flow/skills/reanchor/evals/evals.json:76` — eval case `id: 7`'s `name` field, `negative-routing-rule-discipline-is-re-anchor-plugin` -> `negative-routing-rule-discipline-is-discipline-plugin`, matching the sibling `negative-routing-*` case names. `expected_output`/`expectations` were already correct; only the `name` field was stale. Grepped the repo — nothing else references the old name. Each plugin touched (`work-items`, `session-flow`) gets a matching patch version bump and CHANGELOG entry. **Item 1 of #1328 (orphaned `pluginConfigs` overlays) is deliberately NOT addressed — its premise is false, verified against current upstream docs.** #1328 claimed the rename orphans `pluginConfigs["re-anchor@melodic-software"].options` because the `renames` map "migrates only the enabled/installed record — not the options," and proposed telling users to manually re-enter their four `userConfig` values. Fetched [Create and distribute a plugin marketplace](https://code.claude.com/docs/en/plugin-marketplaces#rename-or-remove-a-plugin) this session; it states, verbatim: > It then rewrites the old key to the new key in the user, project, and local settings scopes for both `enabledPlugins` and `pluginConfigs`, so the notice appears once. `plugins/discipline/CHANGELOG.md`'s `[0.9.0]` entry already states this correctly — a `chatgpt-codex-connector[bot]` reviewer raised the same pluginConfigs concern on #1276 itself, and it was resolved pre-merge (not left "without a reply" as #1328 states): confirmed incorrect against the same upstream doc and fixed in commit `aa094fc9`, before #1276 merged. Adding the proposed "re-enter your values via `/plugin configure discipline`" line would ship guidance that contradicts upstream behavior. Per the Bug Investigation Rule (`/work-items:work`), this is proof via history that the claim doesn't apply, not a reproduction — so item 1's acceptance box is intentionally left unchecked. The three retained-by-design items #1328 lists (the `re-anchor-audit-correct.md` filename, `marketplace.json` `tags`/`renames`, CHANGELOG history) are unchanged, per the issue's own acceptance criterion 3. ## Test plan - [x] `jq .` on all three touched JSON files (`evals.json`, both `plugin.json`) — valid. - [x] Grepped the fixed lines post-edit — no `re-anchor` token remains at any of the three sites; grepped the eval-case-name string repo-wide for other references — none found. - [x] Diffed the retained-by-design sites (`plugins/discipline/context/re-anchor-audit-correct.md`, `.claude-plugin/marketplace.json`, `docs/topics/fresh-eyes-checkpoint-audit/PLAN.md`) — untouched. - [x] Verified the item-1 rebuttal against current docs (WebFetch, this session) and against PR #1276's own review thread (`gh api repos/.../pulls/1276/comments`). ## Related Closes #1328. Related to #1276 (the rename this fixes fallout from). --- *This was generated by AI during work-loop execution.* 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-authored-by: Claude Sonnet 5 <noreply@anthropic.com>
The contract tier is committed on the task branch only and pruned before merge (docs/conventions/topic-docs/README.md:33, :302-304). Its durable outcome graduated to docs/adr/0004 in an earlier commit on this branch; what remains here is the evidence base that produced it. Pruned: the thirteen blind section digests, the agent brief, the collision register, and the source-article capture. Each is dated measurement taken against a tree that has since moved — #1276 renamed the re-anchor plugin and the sweep-all-disciplines skill out from under every path they cite — so retargeting them would fabricate precision they never had, and keeping them would publish stale targets an incumbent search silently fails against. The rename mapping and the digests' surviving conclusions are recorded in the ADR. The evidence does not become unreachable. This commit's parent holds the full slice, and GitHub serves blobs at a pull request's head SHA after squash-merge and branch deletion (verified against merged #1365). The PR description carries that SHA and the retrieval command, which is the pointer this convention directs durable surfaces to leave (:135-141, :302-304). Co-authored-by: Claude Opus 5 (1M context) <noreply@anthropic.com>
…dispatched instruction-surface pass (#1322) The design record behind #1315, #1316, and #1318 — the decomposition of the source article, the proportionality gate that decided what to build, and the contracts the shipped code implements. No linked issue. ## What it decides Seven candidate detectors (D1–D7) went through a proportionality gate. **One survived as a new check** — cross-surface instruction conflict, the only finding with no incumbent. Every other candidate became an edit to a check that already owns its surface, or an explicit exclusion with a recorded reason. The gate's own working is here, including the parts that went against the first answer. ## The corrections are the point This branch documents four occasions where the effort asserted "nothing covers this" and was wrong each time, because nobody read the body of the incumbent — `skill-quality:check`, `mcp-tools:audit`, `claude-memory`'s C6 (on the deliverable's *own* primary check, *after* the failure mode had been named), and an eval runner that already existed as working code. Counts had the same disease. The standing rule that came out of it — derive counts by command, never transcribe them — is recorded with the instances that produced it. ## Independent verification A second derivation of the same source article was made **deliberately blind** to this one, and the two were compared. They substantially agree: the blind reader independently reached "one detector with no incumbent, everything else an edit inside the plugin that already owns the surface", and independently routed its own new rule to the same catalog at the same check number. Six findings moved a disposition and are recorded, along with seven things this derivation found that the blind one missed. Where they genuinely diverge — the blind derivation ratified "no new router" — the disagreement is recorded rather than resolved in this work's favour, with the reconciliation left open as a row due before Phase 9. ## A convention conflict this PR cannot resolve on its own `docs/conventions/topic-docs/` places `docs/topics/<slug>/` in the **contract tier**: "committed on the task branch only; pruned before merge." This PR does not prune, because **fifteen topic directories are already on `main`**, including one merged today. Practice and the written convention disagree wholesale, and this branch follows practice rather than silently picking a side. Reviewer call: merge as-is (consistent with every prior topic branch), or prune first (consistent with the written convention, which would then need every existing topic directory explained). Either way the convention and practice should stop disagreeing. Task #37 was resolved on the assumption the prune is real, and that work stands regardless — the shipped catalog carries its own source URLs and recheck triggers rather than pointing into this tree, so nothing in #1316 or #1318 dangles whichever way this goes. ## Also in here The prompt-injection threat model the design entirely lacked — `audit-pass` reads instruction surfaces that are attacker-controlled text in any repository the operator did not write. Plus the re-run contract, the determinism tiering, the seam resolution, and the naming record for `audit-pass` (32 candidates, five lenses, with what the chosen name *costs* recorded so it is not re-litigated). Rebased onto current `main`, including the `re-anchor` → `discipline` plugin rename from #1276. ## Related - #1315 — cross-plugin hygiene, the fixes this design's audit found - #1316 — the criteria payload - #1318 — the `audit-pass` skill - #1096 — fresh-eyes doctrine and check 21; this design's determinism contract was re-verified against it - #1225 — a second sweep-shaped effort; reconciliation open before Phase 9 🤖 Generated with [Claude Code](https://claude.com/claude-code) --------- Co-authored-by: Claude Opus 5 (1M context) <noreply@anthropic.com>
…ferences (#1386) Closes #1283 ## Summary Six stale references survived three sweep passes during the `re-anchor` → `discipline` plugin rename (#1276). All six were two syntactic shapes `rename-references`' pattern library did not cover, and the gap is structural rather than incidental: **Forms 1–12 all assume the renamed token is a skill or mode identifier.** When a CONTAINER renames — a plugin, a marketplace entry — the token occupies positions none of them reach: - **Form 1** anchors on `/<old>`, so it cannot fire on `/plugin configure <old>`: the slash belongs to `plugin`, and the token sits downstream in argument position. - **Form 3** needs a path; none of these are paths. - **Form 2** (bare token) matches, but cannot separate the container sense from the verb sense at any triage setting when the token is also a verb in the consuming codebase. Forms 13–15 anchor on syntax that admits **only** the naming sense: | Form | Position | Why it is Certain | |---|---|---| | 13 | Command argument — `/plugin install <old>@mkt`, `/plugin configure <old>` | A management verb immediately precedes the token; prose does not say "/plugin configure" before an English verb | | 14 | Document title — `# <old>`, frontmatter `name:` | The `$` anchor: a heading that *contains* the token may be verb usage, but one that IS the token can only be naming it | | 15 | Possessive / appositive — `<old>'s`, `the <old> plugin` | English verbs do not take the possessive clitic; the noun-class appositive forces the naming reading | Command-argument hits are flagged as **functional breaks**, not cosmetic: a reader following `/plugin install <old>@marketplace` gets `plugin-not-found`. Four of the six missed references were this shape, including the README's own install block. ## Fix - `context/patterns.md` — Forms 13–15, each with the five documented fields the existing forms carry, under a short section explaining why container position is its own class. - `context/triage.md` — records the collision class the English-verb blocklist **cannot** serve. The blocklist holds tokens that are verbs *in general*; a token that is a verb *in the consuming codebase* fails both ways: omitted → every bare-token hit is rated Certain and the sweep proposes rewriting the verb uses; added → every hit lands ambiguous, where the per-match confirmation rule turns a handful of real defects into hundreds of prompts. Extending the blocklist swaps one unusable bucket for another. The remedy is position. - `context/patterns.md` Phase 6 — now requires validating any new form on **both** axes. Recall alone is not evidence: Form 2 already has perfect recall on every form in the library and is still unusable. - `context/audit.md` — pattern-form breakdown lists 13–15, so an audit report accounts for every form the sweep runs. - Seven eval cases (7-13): the container-rename sweep, and the blocklist-extension trap. ## Verification Validated on **both** axes against the real fixture rather than asserted. Recall came from the removed lines of `930c97a4` (the commit that fixed the references — its deletions *are* the defect set); precision from the whole pre-fix tree at `930c97a4^`. | Pattern | Hits on `930c97a4^` under `plugins/discipline` | Real defects among them | |---|--:|--:| | Form 2 (bare token) | **134** | 8 | | Forms 13–15 combined | **9** | 8 | The 9th hit is a frozen CHANGELOG-history line, which the existing "Frozen historical records" auto-exclusion already handles. The one defect line Forms 13–15 do *not* match is the frontmatter `description` trigger-phrase block — deliberately kept in that rename, so matching it would have been a false positive. Gates run locally: - `claude plugin validate .` — passes - `markdownlint-cli2` on all four changed markdown files — 0 errors - `scripts/check-changelog-parity.sh --check-bump origin/main` — passes (`docs-hygiene` 0.8.6 → 0.9.0 with a matching entry) - `scripts/check-changelog-parity.sh --check` — passes - `evals.json` parses; diff is additive only (25 insertions, 0 deletions to existing cases) ## Review rounds (this PR supersedes #1335 for branch reasons only — same work, rebased) Four findings across two reviewers, all verified against real input before fixing; two were correct about defects in **this PR's own claims**: - **P1 — Form 2 was never suppressed.** The first version added Forms 13–15 as the remedy for a codebase-specific verb, but nothing suppressed Form 2, so every prompt they were meant to avoid still fired. Fixed by container-position precedence: dedup by `(file, line)` after the sweep, before triage. - **P1 (second round) — precedence left the residue.** Dedup only resolves lines a container form ALSO matched (8 of 134 on the fixture). The other 126 still fell through to Form 2's Certain default, so my claim that they "were never candidates" was false. Fixed by container-rename mode: mode is decided by WHAT is renamed, so it works where the static blocklist cannot. - **P2 — Form 14 over-reached on ordinary-word names.** Verified in this repo: renaming a `testing` plugin matches `README.md:86` (`### Testing`), renaming `architecture` matches `plugins/miro/README.md:39`. Under precedence a false Certain there is worse than a Form 2 hit — it discards the safer classification. Form 14 is now scoped to container-owned files and always Ambiguous for common-word tokens. - **P2 — regex gaps.** Form 14 missed single-quoted YAML; Form 15 missed the token as inline code before the clitic (its own motivating example). Both fixed with paired-quote alternation and optional backticks, verified against real input including the mismatched-quote rejection case. Final fixture result: **8 Certain findings, 126 reported-not-proposed, 0 confirmation prompts**, against Form 2 unaided at 134. ## Related - Refs #1394 — filed from this PR's review: Form 14 is case-sensitive, so a title-cased heading falls through to Form 2 and gets a LESS safe classification. Deliberately not fixed here — the safe version needs the scope rule load-bearing first. - Refs #1335 — the original PR for this work; closed unmerged because a force-push was needed after main moved and force-push is blocked in this environment. Identical content, rebased. - Refs #1276 — the rename whose six missed references are this PR's fixture; `930c97a4` is the reference commit 🤖 Generated with [Claude Code](https://claude.com/claude-code) https://claude.ai/code/session_01GSXnCLnmzk8y4cKv2y1z9f --------- Co-authored-by: Claude Opus 5 (1M context) <noreply@anthropic.com>
1 participant
No related issue: the rename was decided in-session; the follow-on metadata work it unblocks is tracked under the issues in ## Related.
Summary
Renames the
re-anchorplugin todiscipline.re-anchoris a bare verb sitting among 60 noun-named siblings, itsre-prefix presumes a prior anchored state the conversation-start case does not have, and it collides with the cognitive-bias term of art.disciplineis the word the plugin already used for itself 189 times ("a drift corrector for research discipline") — the authors' own revealed vocabulary, not a coinage, and legible to a reader who has never seen the plugin.The verb "re-anchor" is deliberately KEPT (210 occurrences,
context/re-anchor-audit-correct.md, and 3 live trigger phrases). The three defects above concern its use as a CONTAINER name; as the verb for step 1 of the loop it is accurate. Changing it would touch eval expectations and trigger keywords thatskill-quality:checkgates against HEAD, so it belongs in a separate reviewable pass if wanted at all.Mechanically, 45 files:
git mvso history is preservedsweep-all-disciplinesskill renamed tosweep-allplugin.jsonname + version 0.9.0; marketplace entry name/source plus arenamesmap entry so installed consumers migrate instead of hittingplugin-not-found/re-anchor:*invocations to/discipline:*;metadata.re-anchor-batchtodiscipline-batchdocs-hygiene,planning,review,session-flow,source-control,work-itemsDeliberately NOT rewritten: CHANGELOG history and
docs/topics/— they record releases and decisions that genuinely happened under the former name.Test plan
claude plugin validate .— passes; therenameschain terminatesmarkdownlint-cli2over all 26 changed markdown files — 0 errorsgit grep -i 're-anchor'— 0 stale/re-anchor:invocations and 0 staleplugins/re-anchorpaths outside CHANGELOG history anddocs/topics/; every remaining hit is the verb, verified file by filecode-tidying,context7,playwrightvendor, anddocs-hygienedoc examples; untouched, not ours)plugins/re-anchor/orplugins/discipline/. Trivial single-file overlaps only: feat(x): add X (Twitter) to Markdown plugin #1263 (marketplace.json,README.md), docs: answer the loop-engineering questions from the corpus and land the sweep's corrections #1266 (docs/conventions/loop-lane/README.md), feat(skill-quality): fresh-eyes delegation doctrine + conformance gate (check 21) #1096 and feat(session-flow): add find-handoff skill to recover a lost handoff (#976) #1033 (README.md)origin/mainatd39ea667; cleanRelated
renamesmap entry stops being requiredsweep-all-disciplinestosweep-alldiscipline(do-your-research)after this mergesrename-referencespattern library missed the two syntactic shapes that produced the six stale references fixed in930c97a4🤖 Generated with Claude Code
https://claude.ai/code/session_01GSXnCLnmzk8y4cKv2y1z9f