agentleFS
jellydnjellydn/my-ai-tools/skills/doc-search/SKILL.md

doc-search

Skill127 starsChanged 27 days agoMITView on GitHub

Search project documentation — ADRs, wiki entries, conventions via ripgrep, qmd, fff, and ctx

2 other projects use a version of this.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

11 rules. About 0 are common, 7 are distinctive, and 4 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈7

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 4

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Use this skill when:

    line 14
  2. Very specific:

    Use fff to locate documentation files:

    line 45
  3. Very specific:

    Use rg (ripgrep) to search documentation content:

    line 54
  4. Very specific:

    Use qmd to search durable project knowledge:

    line 64
  5. Very specific:

    Use ctx to find previous discussions about documentation:

    line 76
  6. Distinctive:

    Before writing new documentation, study existing examples:

    line 115
  7. Distinctive:

    Match conventions: Use the same formatting patterns

    line 120
  8. Distinctive:

    docs-update: Use doc-search first to find what exists before updating

    line 151
  9. Distinctive:

    qmd-knowledge: Add new knowledge to durable storage

    line 154
  10. Distinctive:

    Search before writing: Always check if documentation already exists

    line 159
  11. Distinctive:

    Update qmd: After finding useful documentation, index it in qmd

    line 162

≈ “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.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Documentation Search

When to Use

Use this skill when:

  • Starting work on a feature that may have existing decisions or ADRs
  • Need to understand project conventions or patterns
  • Looking for past design discussions or trade-off analyses
  • Want to find relevant wiki entries or knowledge base articles
  • Need to understand how existing documentation is structured
  • Planning to write documentation and want to match existing style

What It Does

Helps you find and navigate existing project documentation using available tools (grep, qmd, fff, ctx). Documentation is often scattered across multiple locations — this skill provides a systematic approach to find what you need.

Where Documentation Lives

This project stores documentation in several locations:

LocationContentBest For
docs/User-facing guides (quick-start, tutorials)Getting started, feature guides
ADRs via configs/adr/*Architecture Decision RecordsWhy decisions were made
wiki/LLM wiki — entities, concepts, logKnowledge base, cross-references
MEMORY.mdDurable project learnings and gotchasKnown issues, conventions
skills/*/SKILL.mdAgent skill definitionsTool capabilities
configs/*.mdTool-specific documentationPer-tool behavior
AGENTS.mdAgent instructions and guidelinesDevelopment workflow
qmd knowledge baseIndexed project knowledgeSearchable, AI-powered retrieval

How to Search

Step 1: Find Relevant Documents

Use fff to locate documentation files:

fff "*.md" docs/         # All documentation files
fff "*adr*"              # Architecture Decision Records
fff "*auth*" docs/       # Auth-related documentation
fff "*wiki*"             # Wiki entries

Use rg (ripgrep) to search documentation content:

rg "decision" docs/                   # Find decisions in docs
rg "ADR-0" . --glob "*.md"           # Find ADR references
rg "rate.limiting" --glob "*.md"     # Find docs about rate limiting

Step 2: Query Knowledge Base

Use qmd to search durable project knowledge:

qmd search "authentication decisions"     # Find related knowledge
qmd query "What architecture exists for X?"  # Get structured answers
qmd get ADR-001                           # Get a specific ADR

The qmd knowledge base indexes project learnings, ADRs, conventions, and gotchas that persist across agent sessions.

Step 3: Search Past Sessions

Use ctx to find previous discussions about documentation:

ctx search "documentation" "decision"       # Past doc discussions
ctx search "ADR" "architecture" path/docs/  # Past ADR discussions
ctx search "convention" "pattern"            # Past convention discussions

Step 4: Explore by Pattern

Project documentation follows consistent patterns:

Architecture Decision Records (ADRs):

# Find all ADRs
ls -la docs/adr/
# Search ADR content
rg "decision" docs/adr/

Wiki entries:

# Browse wiki structure
ls -la wiki/wiki/entities/
ls -la wiki/wiki/concepts/
# Search wiki content
rg "topic" wiki/wiki/ --glob "*.md"

Agent skills:

# Find skill documentation
rg "what.*does" skills/*/SKILL.md
# Find skills by category
rg "compatibility:.*auth" skills/*/SKILL.md

Understanding Documentation Style

Before writing new documentation, study existing examples:

  1. Pick a reference: Find an existing doc on a similar topic
  2. Analyze structure: Note heading levels, code blocks, tables
  3. Check tone: Is it formal, conversational, or technical?
  4. Match conventions: Use the same formatting patterns

Example: If you need to write an ADR, read an existing ADR first:

fff "*adr*.md"   # Find existing ADRs
cat docs/adr/001-some-decision.md  # Read as reference

Documentation Discovery Workflow

Before writing:
1. Search existing docs for related content
2. Check qmd for existing knowledge
3. Search ctx for past discussions
4. Read existing docs for style reference

During writing:
5. Link to relevant ADRs and decisions
6. Reference existing conventions
7. Cross-link related documentation

After writing:
8. Add to wiki or qmd knowledge base
9. Reference from relevant agent instructions
10. Update docs-update skill if patterns changed

Integration with Other Skills

  • docs-update: Use doc-search first to find what exists before updating
  • context-discovery: Documentation search is part of context discovery
  • implementation-logger: Extract learnings into documentation
  • qmd-knowledge: Add new knowledge to durable storage
  • llm-wiki: Build and maintain wiki entries from findings

Tips

  • Search before writing: Always check if documentation already exists
  • Follow existing patterns: Match the style of existing docs
  • Cross-link: Reference ADRs, wiki entries, and related docs
  • Update qmd: After finding useful documentation, index it in qmd
  • Use rg over grep: ripgrep is faster and respects .gitignore
  • Check wiki/ first: The LLM wiki is designed for discoverability

Download now (ZIP)

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
What did you use it for? e.g. Reviewing pull requests in a Rails app

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.