skills.rest
openJiuwen-ai
openJiuwen-aiOFFICIAL
Explore all skills by this author →
WebsiteGitHub

project-maintainer

Builds and maintains a structured in-repository project knowledge base with symbol-level coverage and health audits.

7.1k|1.1k|Updated Mar 5, 2026
Search Tags:#Documentation#code health#repository mapping#symbol audit#Knowledge Base#Multi-Agent#Code Analysis
One-click install
npx skills add https://github.com/openJiuwen-ai/jiuwenswarm --skill project-maintainer
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: project-maintainer
Source: https://github.com/openJiuwen-ai/jiuwenswarm/tree/main/jiuwenswarm/resources/agent/workspace/skills/project-maintainer
Command: npx skills add https://github.com/openJiuwen-ai/jiuwenswarm --skill project-maintainer

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) components.

What problem does it solve?

Keeping project documentation accurate as code evolves is hard: docs drift out of sync, large repositories are never fully mapped, and there is no reliable record of which classes, functions, and methods have been reviewed for health and risk. This Skill creates and maintains a structured, agent-readable knowledge base inside the target repository at .doc_project_maintainer/ so future developers and agents can query modules, directories, cross-boundary flows, and every top-level class, function, and method without rediscovering the codebase.

Core Features & Use Cases

  • Structured Project Mapping: Documents modules, directories, cross-boundary flows, decisions, and change records with a manifest, index, and build plan, kept small enough for selective reading.
  • Exhaustive Code Symbol Coverage: Inventories every stable source file, top-level class, top-level function, and class method using scripts/inventory_symbols.py, generating coverage maps and symbol audit maps with hash-based staleness detection.
  • Symbol Health Audits with Integrity Controls: Records evidence-based health dimensions and issues per symbol, with HMAC-signed audit promotion, verification, and an HTML audit report via scripts/audit_integrity.py and scripts/render_audit_report.py.
  • Use Case: Ask an agent to analyze a large legacy repository. The Skill builds the artifact in slices, coordinates multi-agent coverage, audits runtime symbols for risk, and keeps the docs synchronized after every verified bug fix or refactor.

Quick Start

Ask the agent to analyze this repository and build the project-maintainer knowledge base with full code symbol coverage.

Frequently Asked Questions about project-maintainer

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I document an entire repository with an AI agent?▼

Run the project-maintainer workflow: it explores the repository, inventories every stable source file and top-level symbol with scripts/inventory_symbols.py, then builds module, directory, flow, and code symbol docs in slices under .doc_project_maintainer/ until the coverage closure audit passes.

How to audit code health for every function and method in a project?▼

Use the symbol health audit workflow: each top-level class, function, and method gets an entry doc with health dimensions and issues, then a real audit agent reviews the implementation and promotes the record through scripts/audit_integrity.py before it counts as trusted.

Does the symbol inventory script require external dependencies?▼

No, scripts/inventory_symbols.py runs on a plain Python environment with no external dependencies. It uses a built-in Python AST extractor, optionally uses ctags when available, and falls back to a dependency-free heuristic extractor for other languages.

How does the skill keep documentation in sync after code changes?▼

It compares file and symbol hashes recorded in coverage-map.json and symbol-audit-map.json against current source, marks changed symbols stale or audit-expired, and synchronizes only the affected artifact slices after verified fixes. Python symbols use normalized AST hashes so formatting changes do not expire audits.

What are the limitations of heuristic symbol extraction?▼

Heuristic extraction can scaffold documentation but carries lower confidence and may miss symbols in unsupported languages. Files flagged heuristic or requires_review keep coverage partial until manually reviewed, re-run with a stronger extractor, or explicitly marked out of scope.

Related Skills

Explore More Software Engineering Skills→
davepoondavepoon
3.4k

gsd:new-workspace

Create isolated workspaces with git worktrees or clones and independent .planning directories.

Community
Intermediate
yulonghe97yulonghe97
8

docs

Update project documentation to reflect completed, verified work.

Community
Intermediate
Ramprasad4121Ramprasad4121
1

audit-prep-assistant

Prepare codebases for security review with static analysis and documentation.

Community
Advanced
globalmsqglobalmsq

moai-docs-generation

Automate documentation generation across codebases with Sphinx, MkDocs, and TypeDoc.

Official
Intermediate