wiki-doc
Verified3-level documentation protocol for LLM-assisted codebases. Builds and maintains a structured knowledge base (module index → function index → function docs) that any LLM agent can navigate with minimal context loading. Supports two pass types: doc-only passes (no source edits) and edit-enabled passes (small fixes surfaced during documentation, gated by an explicit suggest-approve-apply-verify-log protocol). Use this skill whenever the user wants to: create documentation for a codebase or module, build a knowledge base from code review sessions, set up structured docs that stay synchronized with source code, run a "salvage" documentation pass on existing reviewed code, maintain or refresh existing wiki-doc documentation after code edits, trace variable/argument relationships across modules, generate AST-based function indexes, or audit a module for duplicates and dead code before documenting it. Trigger phrases: "wiki-doc", "3-level docs", "function index", "salvage documentation", "Level 2 index", "Level 3 docs", "document this module", "create the variable map", "refresh line numbers", "audit this module before documenting", "edit-enabled doc pass".
| File | Kind | Size |
|---|---|---|
| SKILL.md | skill md | 61.8 KB |
| examples/README.md | markdown | 3.2 KB |
| examples/glycomsp_companion.md | markdown | 11.5 KB |
| scripts/ast_index.py | script | 17.5 KB |
examples/README.md
View on GitHub# Examples
This folder collects real project companion files showing how concrete project-side
content (paths, verification rules, naming conventions, bug-file taxonomies) layers
on top of the generic wiki-doc skill.
The skill itself (`SKILL.md` at repo root) stays project-agnostic. Each project that
uses wiki-doc maintains its own companion file describing the project's specific
conventions and configuration. New projects crib from existing companions when
authoring their own.
## What's here
- **[`glycomsp_companion.md`](glycomsp_companion.md)** — GlycoMSP, the project that
the v1.0 protocol matured against. ~10,000-line Python module documented across
multiple sessions in April–May 2026. Eight protocol gaps were surfaced and
resolved during the salvage; the companion documents the project-side
configuration plus the v0.2 → v1.0 phase numbering migration.
## How to read a companion file
Companion files have a typical shape:
1. **Migration note** — only if the project predates the current skill version, to
document any breaking changes (e.g., phase numbering shifts).
2. **Pass-type mapping** — if the project's CLAUDE.md (or equivalent project-wide
convention) defines its own editing modes, map them to the skill's neutral
terminology.
3. **Project config block** — full YAML with project-specific values. Acts as a
real-world example of what to put in `.claude/wiki-doc.config.yaml`.
4. **File-naming conventions and worked examples** — the project's bug-file
format, dead-code organization, line-delta tracking shape, etc.
5. **Source-internal metadata** — what fields the project maintains, how they're
bumped.
6. **Section terminology mapping** — disambiguates any in-project terms that
overlap with skill terms.
7. **Sub-agent usage** — any project-specific dispatch patterns.
8. **Salvage history reference** — timeline of when each phase was completed for
the project's modules.
The goal is to make the project legible to a fresh agent reading the skill: SKILL.md
teaches the protocol, the companion supplies the concrete instances.
## Authoring your own companion
When your project's wiki-doc conventions stabilize (typically after 2–3 modules have
been salvaged), it's worth writing a companion file. A typical workflow:
1. **Copy `glycomsp_companion.md`** to your project's docs directory and rename it
`<your-project>_companion.md`.
2. **Strip the GlycoMSP-specific content** down to the section structure.
3. **Fill in your project's** paths, config values, naming conventions, and any
project-specific patterns.
4. **Optional — submit a PR** to add your companion as another worked example. See
[`CONTRIBUTING.md`](../CONTRIBUTING.md) for the flow. Real-world examples are
the most valuable contributions to this folder.
## What companion files are NOT
- **Not a fork of SKILL.md.** The companion layers on top; it doesn't replace
protocol content.
- **Not a tutorial.** That's [`docs/quickstart.md`](../docs/quickstart.md).
- **Not a substitute for project-internal docs.** A companion file is the bridge
between the skill and the project's wiki-doc'd output. Other project docs
(architecture, design, runbooks) are unaffected by wiki-doc and live wherever
they normally would.
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| rea-analysis-optimizationmorluto/reaOptimize REA providers and shared analysis workflows for agent context, latency, CPU, and memory while preserving correctness and Evidence. Use for analysis bottlenecks and cache or query design; CI/test-suite optimization belongs to rea-ci-tests-audit. | Stars: 471k | Updated | Claude CodeCodex | Verified | Install |
| rea-ci-tests-auditmorluto/reaAudit REA CI and test suites for correctness, fixture drift, false passes, and measured performance; use for suite or workflow investigations and efficiency changes, rather than ordinary feature testing or a single test failure. | Stars: 471k | Updated | Claude CodeCodex | Verified | Install |
| rea-tool-designmorluto/reaAudit, investigate, design, or change REA investigation tools, CLI/MCP contracts, provider capabilities, and Evidence semantics. | Stars: 471k | Updated | Claude CodeCodex | Verified | Install |
| ponytail-auditDietrichGebert/ponytailQuality audit of the whole repo: bugs, security, real load, missing tests, speed, and what to delete. Most important first. | Stars: 194k | Updated | Claude CodeCodex | Verified | Install |