wiki-refactor
Skill8 starsChanged 6 months agoMITView on GitHub
Conductor skill for full wiki refactoring. Orchestrates a 7-phase pipeline — discovery, deduplication, information architecture, writing plan, progressive rewrite + review, quality metrics, and safe delivery. Enforces PRD protection (hard gate), human checkpoint after planning, scope caps, and content snapshot/drift detection.
1 other project uses a version of this.
Adopted by
Most recently taken up by
- dexio-wiki/llm-wiki-skillstheir own version, first seen Sep 28, 2026
Other versions1 project uses a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 1
- dexio-wiki/llm-wiki-skills★ 1same name
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository129 other files it gives its agents.
- AGENTS.mdAGENTS.md
- guidance/AGENTS.mdAGENTS.md
- CLAUDE.mdCLAUDE.md
- guidance/CLAUDE.mdCLAUDE.md
- .github/copilot-instructions.mdCopilot instructions
- blast-radius-checkSkill
- brainstormingSkill
- branch-flow-gateSkill
- branch-sync-gateSkill
- codebase-reconSkill
- codeowners-drift-auditSkill
- code-review-batterySkill
- cognitive-complexity-refactoringSkill
- debateSkill
- debug-conductorSkill
- dispatching-parallel-agentsSkill
- domain-buildSkill
- evidence-adjudicatorSkill
- executing-plansSkill
- external-cli-auditSkill
- feature-developmentSkill
- field-rename-verificationSkill
- finishing-a-development-branchSkill
- git-branch-conventionsSkill
- gitlab-cliSkill
- hotfix-charterSkill
- implementation-trackerSkill
- incident-revert-reviewSkill
- infra-config-investigatorSkill
- investigation-stateSkill
- kernel-splitSkill
- llm-behavior-investigatorSkill
- llm-skill-reviewSkill
- merge-authorization-gateSkill
- micro-harsh-reviewSkill
- output-verificationSkill
- pre-commit-gateSkill
- progressive-code-review-gateSkill
- progressive-harsh-reviewSkill
- providing-code-reviewSkill
- pr-triage-gateSkill
- push-authorization-gateSkill
- receiving-code-reviewSkill
- reproduction-experiment-investigatorSkill
- requesting-code-reviewSkill
- requirements-validation-pmSkill
- requirements-validationSkill
- scope-tripwireSkill
- session-handoffSkill
- skills-hierarchy-tuningSkill
- sp-bughuntSkill
- state-consistency-investigatorSkill
- subagent-driven-developmentSkill
- systematic-debuggingSkill
- test-driven-developmentSkill
- timeline-trace-investigatorSkill
- token-estimationSkill
- unified-commit-gateSkill
- using-git-worktreesSkill
- using-superpowersSkill
- The first 60 shown.
The file
We split this skill into the rules it gives an agent
2 rules. About 0 are common, 1 is distinctive, and 1 is very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈1
Distinctive · estimate
This author's own choices. Worth a look.
1
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 33
When asked to do a full-wiki refactor: eliminate duplication, improve information architecture, or elevate prose quality across many pages
- Very specific: line 93
Run use-skill link-verification (exit 0 required) and use-skill wiki-secret-audit (empty findings)
≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
wiki-refactor — conductor
Full-wiki refactor: eliminate duplication, improve IA, elevate prose quality.
HARD gate with 3 PRD checkpoints and a human checkpoint after Phase 4.
Wrong skill? Single-page edit → wiki-orchestrator · Content coherence →
wiki-content-coherence · Link check → link-verification. Background:
rationale.md.
When to Use
- When asked to do a full-wiki refactor: eliminate duplication, improve information architecture, or elevate prose quality across many pages
- Triggered by:
refactor wiki,wiki overhaul,reorganize wiki pages,wiki:refactor,reorganize all wiki pages - Wrong skill? Single-page edit →
wiki-orchestrator· Page-level coherence →wiki-content-coherence
🔴 PRD protection (non-negotiable)
Detect PRD pages (case-insensitive): filename/title contains PRD; body H1/H2
exactly PRD; body header Product Requirements Document; or operator include
list. Detection → HALT. Operator must re-invoke with PRDs scoped out.
| Checkpoint | When | Action |
| PRD-1 | After Phase 1 inventory | Quarantine matches in prd-quarantine-list |
| PRD-2 | Before every Phase 5 write | Match quarantine → HALT |
| PRD-3 | After Phase 7 diff | Any PRD in changeset → HALT |
Pre-flight
: "${WIKI_SEED_URL:?}"
git checkout -b "feat/wiki-refactor-$(date +%F)"
mkdir -p wiki-refactor-artifactsPhases (each references the detailed runbook)
| # | Phase | Runbook | Output |
| 1 | Discovery — crawl descendants, snapshot content | phase-1-discovery.md | wiki-inventory.md |
| 2 | Deduplication — analyze duplication patterns | phase-2-deduplication.md | dedup-analysis.md |
| 3 | Information Architecture — new structure | phase-3-information-architecture.md | new-wiki-structure.md |
| 4 | Writing plan — source→target page map | phase-4-writing-plan.md | wiki-writing-plan.md |
| 5 | Rewrite + 3-round progressive review | phase-5-rewrite-and-review.md | refactored-wiki/*.md |
| 6 | Quality metrics | phase-6-quality-metrics.md | wiki-refactor-report.md |
| 7 | Safe delivery — branch, migration guide, final checks | phase-7-safe-delivery.md | Ready-to-merge branch |
Phase 1 scope cap
100 pages OR >50k words → warn operator + require confirmation. Zero pages → abort with "wiki unreachable or empty." PRD-1 runs on the full inventory.
⏸️ Human checkpoint (after Phase 4)
Present wiki-writing-plan.md (pages to rewrite / merge / delete / create,
word-count delta, PRD quarantine list). Do not proceed to Phase 5 without
explicit operator approval.
Phase 5 write loop
For each target page, before invoking tools/wiki-write.sh:
- PRD-2 check against quarantine list
- 3-round progressive review (use
progressive-harsh-review) - Stage 5.5 structure gate (
node tools/wiki-markdown-validate.js) tools/wiki-write.sh update --doc $ID --content page.md(exit0or halt)
Phase 7 delivery
- Package refactored pages into
wiki/on the feature branch - Emit migration guide (old-slug → new-slug)
- Diff Phase-1 snapshot vs current wiki → on drift, present overwrite / merge / abort
- Run
use-skill link-verification(exit 0 required) anduse-skill wiki-secret-audit(empty findings) - PRD-3 check: final diff contains zero PRD files
Failure modes
| Symptom | Cause | Response |
| PRD checkpoint fires | PRD in rewrite scope | HALT — no recovery path |
| Phase 1 returns zero pages | Bad URL, auth, empty wiki | Abort with diagnostic |
| Phase 2 zero duplicates | No duplication | Continue; Phase 3 may still apply |
| Scope cap exceeded | >100 pages or >50k words | Warn operator; narrow scope |
| Phase 7 snapshot drift | Wiki edited during refactor | Present diff; operator picks overwrite/merge/abort |
| 3-round review won't converge | Quality blocker | Escalate page to human |
Success criteria and PRD-protection rationale: rationale.md.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

