Single source of truth and librarian for ALL Claude official documentation. Manages local documentation storage, scraping, discovery, and resolution. Use when finding, locating, searching, or resolving Claude documentation; discovering docs by keywords, category, tags, or natural language queries; scraping from sitemaps or docs maps; managing index metadata (keywords, tags, aliases); or rebuilding index from filesystem. Run scripts to scrape, find, and resolve documentation. Handles doc_id resolution, keyword search, natural language queries, category/tag filtering, alias resolution, sitemap.xml parsing, docs map processing, markdown subsection extraction for internal use, hash-based drift detection, and comprehensive index maintenance.
Installer avec Codex ou Claude Copiez ce prompt, collez-le dans Codex, Claude ou un autre assistant, puis laissez-le vérifier la page du skill et l'installer pour vous.
Une commande directe contourne le prompt de vérification. Examinez la source avant de l'exécuter.
Single source of truth and librarian for ALL Claude official documentation. Manages local documentation storage, scraping, discovery, and resolution. Use when finding, locating, searching, or resolving Claude documentation; discovering docs by keywords, category, tags, or natural language queries; scraping from sitemaps or docs maps; managing index metadata (keywords, tags, aliases); or rebuilding index from filesystem. Run scripts to scrape, find, and resolve documentation. Handles doc_id resolution, keyword search, natural language queries, category/tag filtering, alias resolution, sitemap.xml parsing, docs map processing, markdown subsection extraction for internal use, hash-based drift detection, and comprehensive index maintenance.
allowed-tools
Read, Glob, Grep, Bash, Skill
Claude Docs Management
🚨🚨🚨 MANDATORY: Hybrid Parallel Pattern 🚨🚨🚨
STOP - Before using this skill for ANY Claude Code documentation query:
IF YOU ARE THE MAIN AGENT, you MUST invoke BOTH sources in the same message:
This skill (docs-management) - local cache, token-efficient
claude-code-guide subagent - live web search
[Skill tool: docs-management]
"Find documentation about {topic}"
[Task tool: claude-code-guide] (SAME MESSAGE - USE THIS EXACT PROMPT)
"First WebFetch https://code.claude.com/docs/en/claude_code_docs_map.md to find
relevant doc pages about {topic}. Then WebFetch those specific pages. Use WebSearch
only if needed for additional context. Do NOT use Skill tool (not available).
Return key findings with source URLs."
claude-code-guide does NOT have Skill tool access. Always prompt it
to use . If you see "No such tool: Skill" error,
you prompted it wrong.
⚠️ CRITICAL:
WebSearch/WebFetch explicitly
This is AUTOMATIC. Do NOT wait for user to ask for it.
IF YOU ARE A SUBAGENT: Note in your response that main agent should also query claude-code-guide.
🚨 CRITICAL: Path Doubling Prevention - MANDATORY
ABSOLUTE PROHIBITION: NEVER use cd with && in PowerShell when running scripts from this skill.
The Problem: If your current working directory is already inside the skill directory, using relative paths causes PowerShell to resolve paths relative to the current directory instead of the repository root, resulting in path doubling.
REQUIRED Solutions (choose one):
✅ ALWAYS use helper scripts (recommended - they handle path resolution automatically)
✅ Use absolute path resolution (if not using helper script)
✅ Use separate commands (never cd with &&)
NEVER DO THIS:
❌ Chain cd with &&: cd <relative-path> && python <script> causes path doubling
❌ Assume current directory
❌ Use relative paths when current dir is inside skill directory
For all scripts: Always run from repository root using relative paths, OR use helper scripts that handle path resolution automatically.
🚨 CRITICAL: Large File Handling - MANDATORY SCRIPT USAGE
⚠️ ABSOLUTE PROHIBITION: NEVER use read_file tool on the index.yaml file
The file exceeds 25,000 tokens and will ALWAYS fail. You MUST use scripts.
✅ REQUIRED: ALWAYS use manage_index.py scripts for ANY index.yaml access:
python scripts/management/manage_index.py count
python scripts/management/manage_index.py list
python scripts/management/manage_index.py get <doc_id>
python scripts/management/manage_index.py verify
All scripts automatically handle large files via index_manager.py.
Available Slash Commands
Use the consolidated docs-ops skill for common workflows:
/claude-ecosystem:docs-ops scrape - Scrape all configured Claude documentation sources, then refresh index and validate
/claude-ecosystem:docs-ops refresh - Refresh the local index and metadata without scraping from remote sources
/claude-ecosystem:docs-ops validate - Validate the index and references for consistency and drift without scraping
/claude-ecosystem:docs-ops rebuild-index - Force rebuild the search index from filesystem
/claude-ecosystem:docs-ops clear-cache - Clear the documentation search cache
Overview
This skill provides automation tooling for documentation management. It manages:
Canonical storage (encapsulated in skill) - Single source of truth for official docs
Subsection extraction - Token-optimized extracts for skills (60-90% savings)
Drift detection - Hash-based validation against upstream sources
Sync workflows - Quarterly maintenance automation
Documentation discovery - Keyword-based search and doc_id resolution
Index management - Metadata, keywords, tags, aliases for resilient references
Core value: Prevents link rot, enables offline access, optimizes token costs, automates maintenance, and provides resilient doc_id-based references.
Quick Reference: Which Command to Use
I want to...
Command
Example
Find docs by keywords
search
search skills progressive
Find docs by natural language
query
query "how to create skills"
Get full document content
content
content code-claude-com-docs-en-skills
Get specific section
content --section
content <doc_id> --section "metadata"
Resolve doc_id to path
resolve
resolve code-claude-com-docs-en-skills
Find related docs
related
related code-claude-com-docs-en-skills
List docs by category
category
category api
List docs by tag
tag
tag skills
Key Workflow:search by keywords → Get doc_id from results → content by doc_id
Important distinctions:
search and query accept flexible keywords (no full doc_id needed)
content, resolve, and related require full doc_id (e.g., code-claude-com-docs-en-skills)
When to Use This Skill
This skill should be used when:
Scraping documentation - Fetching docs from sitemaps or docs maps
Finding documentation - Searching for docs by keywords, category, or natural language
Resolving doc references - Converting doc_id to file paths
Managing index metadata - Adding keywords, tags, aliases, updating metadata
Rebuilding index - Regenerating index from filesystem (handles renames/moves)
Hybrid Parallel Documentation Strategy
CRITICAL: This section is the authoritative source for Claude Code documentation access patterns.
Architecture Constraint
Skills cannot spawn subagents. Only the main conversation thread can use the Task tool.
This means:
When main agent loads this skill → Main agent must also invoke claude-code-guide in parallel
When subagent loads this skill → Subagent should note in response that main agent should invoke claude-code-guide for live coverage
When command invokes this skill → Same as main agent (commands run in main context)
Caller-Aware Guidance
If you are the main agent:
You MUST invoke both this skill AND claude-code-guide subagent in the SAME message
This is the only way to get comprehensive Claude Code documentation coverage
If you are a subagent:
You CANNOT invoke claude-code-guide yourself (architectural constraint)
Include in your response: "For comprehensive coverage, main agent should also query claude-code-guide subagent"
Proceed with local cache results
Always-Parallel Pattern (Main Agent Only)
⚠️ MANDATORY DEFAULT BEHAVIOR - NOT OPTIONAL:
When ANY Claude Code documentation query is detected, the main agent MUST automatically:
Invoke docs-management skill (local cache)
Spawn claude-code-guide subagent (live web) in the same message
This is automatic. The user does NOT need to ask for it.
Detection Triggers
Use both sources automatically when user asks about:
Claude Code features (hooks, skills, memory, MCP, plugins, settings, etc.)
How to do something in Claude Code
Claude Code configuration or troubleshooting
Any topic in the Claude Code documentation
Same-Message Parallel Invocation
# Main agent sends BOTH in a single message (AUTOMATIC):
[Skill tool: docs-management]
"Find documentation about {topic}"
[Task tool: claude-code-guide] (same message = parallel execution)
"First WebFetch https://code.claude.com/docs/en/claude_code_docs_map.md to find
relevant doc pages about {topic}. Then WebFetch those specific pages. Use WebSearch
only if needed for additional context. Do NOT use Skill tool (not available).
Return key findings with source URLs."
IMPORTANT:claude-code-guide is a built-in subagent with tools: Glob, Grep, Read, WebFetch, WebSearch.
It does NOT have the Skill tool - it's designed for web search, not local skill invocation.
Always prompt it to use WebSearch/WebFetch explicitly.
Synthesis Step
After both complete:
Combine results from both sources
Local cache provides category context and token efficiency
Search GitHub issues in anthropics/claude-code for: [ERROR/PROBLEM DESCRIPTION].
Check both open and closed issues. Report issue numbers, status, and any workarounds.
Correct Subagent Pattern
When spawning docs-researcher (or any subagent that uses this skill), the main agent should ALSO spawn claude-code-guide in the same message for comprehensive coverage:
# Main agent spawns BOTH in a single message:
[Task tool: docs-researcher subagent]
"Research Claude Code memory/CLAUDE.md files"
[Task tool: claude-code-guide subagent] (same message = parallel)
"Use WebSearch to find current Claude Code documentation about memory and
CLAUDE.md files on code.claude.com. Return key findings with URLs."
# After both complete, synthesize results
Why both? The docs-researcher uses local cache (fast, token-efficient), while claude-code-guide searches live web (always current). Together they provide comprehensive coverage.
Hierarchical Categories (from docs_map.md)
The index includes category hierarchy from the official Claude Code docs map:
CRITICAL: This section defines HOW to execute operations in this skill.
Delegation Strategy
Default approach: Delegate to Task agent
For ALL scraping, validation, and index operations, delegate execution to a general-purpose Task agent.
How to invoke:
Use the Task tool with:
subagent_type: "general-purpose"
description: Short 3-5 word description
prompt: Full task description with execution instructions
Execution Pattern
Scripts run in FOREGROUND by default. Do NOT background them.
When Task agents execute scripts:
✅ Run directly: python .claude/skills/docs-management/scripts/core/scrape_all_sources.py --parallel --skip-existing
✅ Streaming logs: Scripts emit progress naturally via stdout
✅ Wait for completion: Scripts exit when done with exit code
❌ NEVER use run_in_background=true: Scripts are designed for foreground execution
❌ NEVER poll output: Streaming logs appear automatically, no BashOutput polling needed
❌ NEVER use background jobs: No &, no nohup, no background process management
Anti-Pattern Detection
Red flags indicating incorrect execution:
🚩 Using run_in_background=true in Bash tool
🚩 Repeated BashOutput calls in a loop
🚩 Checking process status with ps or pgrep
🚩 Manual polling of script output
🚩 Background job management (&, nohup, jobs)
🚩 Using BashOutput AFTER Task agent completes ← CRITICAL RED FLAG
If you recognize these patterns, STOP and correct immediately.
After Task Agent Completes
CRITICAL: When the Task agent reports "Done", READ its report and summarize to the user. DO NOT use BashOutput.
Correct workflow:
Task agent reports "Done (X tool uses · Y tokens · Z time)"
✅ READ the agent's message containing its final report
✅ SUMMARIZE the agent's findings to the user
❌ NEVER use BashOutput to "check the real results"
❌ NEVER poll or verify what the agent already reported
Error and Warning Reporting
CRITICAL: Report ALL errors, warnings, and issues - never suppress or ignore them.
✅ Include context: What was being executed when the error occurred
Red flags that indicate issues:
🚩 Non-zero exit code
🚩 Lines containing "ERROR", "FAILED", "Exception", "Traceback"
🚩 "WARNING" or "WARN" messages
🚩 "404 Not Found", "500 Internal Server Error"
Domain-Specific Reporting for Scraping
CRITICAL: When reporting scraping results, distinguish behavior by domain.
Domain-Specific .md URL Behavior:
docs.claude.com and code.claude.com: These domains serve .md URLs successfully
anthropic.com: These domains do NOT serve .md URLs (404 expected, configured with try_markdown: false)
Accurate Reporting:
✅ Good (Domain-Specific): "docs.claude.com: 97 URLs using direct .md (97 skipped/unchanged). anthropic.com: 164 URLs using HTML conversion (158 skipped/unchanged)."
❌ Bad (Misleading): "All .md URL attempts returned 404 (expected - these are HTML pages)" ← This is misleading because Claude domains successfully use .md URLs
Quick Start
Refresh Index End-to-End (No Scraping)
Use this when you want to rebuild and validate the local index/metadata without scraping:
⚠️ IMPORTANT: Use Python 3.13 for validation - spaCy/Pydantic have compatibility issues with Python 3.14+
# Use Python 3.13 for full compatibility with spaCy
py -3.13 .claude/skills/docs-management/scripts/management/refresh_index.py
Optional flags:
# Check for missing files before rebuilding
py -3.13 .claude/skills/docs-management/scripts/management/refresh_index.py --check-missing-files
# Detect drift (404s, missing files) after rebuilding
py -3.13 .claude/skills/docs-management/scripts/management/refresh_index.py --check-drift
# Detect and automatically cleanup drift
py -3.13 .claude/skills/docs-management/scripts/management/refresh_index.py --check-drift --cleanup-drift
This script runs the full pipeline:
✅ Checks dependencies
✅ (Optional) Checks for missing files
✅ Rebuilds the index from the filesystem
✅ Extracts keywords & metadata for all documents
✅ Validates metadata coverage
✅ Generates a summary report
✅ (Optional) Detects and cleans up drift
Expected runtime: ~20-30 seconds for ~500 documents
Scrape All Sources (Used by /scrape-official-docs)
Use this when the user explicitly wants to hit the network and scrape docs:
Python Version Requirement
Python 3.14 works by default for scraping
Python 3.13 required for spaCy operations (keyword extraction, metadata extraction)
# Step 1: Scrape documentation (Python 3.14+ works)
python .claude/skills/docs-management/scripts/core/scrape_all_sources.py \
--parallel \
--skip-existing
# Step 2: IMMEDIATELY run validation after scraping completes# ⚠️ Use Python 3.13 for validation (spaCy compatibility)
py -3.13 .claude/skills/docs-management/scripts/management/refresh_index.py
# Step 3: Clean up aged-out Anthropic articles (reads max_age from sources.json)
python .claude/skills/docs-management/scripts/maintenance/cleanup_old_anthropic_docs.py --execute
Validation and Cleanup Are Required After Scraping
Since --auto-validate is now default: False (for speed), you MUST run validation and cleanup separately immediately after scraping.
Optional: Detect and cleanup drift after scraping:
# Auto-cleanup workflow (detect and cleanup in one flag)
python .claude/skills/docs-management/scripts/core/scrape_all_sources.py \
--parallel \
--skip-existing \
--auto-cleanup
# Then validate (use Python 3.13)
py -3.13 .claude/skills/docs-management/scripts/management/refresh_index.py
Find Documentation
⚠️ CRITICAL: Global flags MUST come BEFORE the subcommand:
# ✅ CORRECT - global flags before subcommand
python find_docs.py --json --limit 10 search skills frontmatter
# ❌ WRONG - global flags after subcommand (will fail with "unrecognized arguments")
python find_docs.py search skills frontmatter --json
Examples:
# Resolve doc_id to file path
python .claude/skills/docs-management/scripts/core/find_docs.py resolve <doc_id>
# Search by keywords (default: 25 results)
python .claude/skills/docs-management/scripts/core/find_docs.py search skills progressive-disclosure
# Search with custom limit (global options come before subcommand)
python .claude/skills/docs-management/scripts/core/find_docs.py --limit 10 search skills
# Search without limit (returns all matching results)
python .claude/skills/docs-management/scripts/core/find_docs.py --no-limit search skills
# Search with minimum score threshold (filters low-relevance results)
python .claude/skills/docs-management/scripts/core/find_docs.py --min-score 20 search skills
# Natural language search
python .claude/skills/docs-management/scripts/core/find_docs.py query "how to create skills"# List by category
python .claude/skills/docs-management/scripts/core/find_docs.py category api
# List by tag
python .claude/skills/docs-management/scripts/core/find_docs.py tag skills
Search Accepts Flexible Keywords (no full doc_ids needed):
Stop words removed automatically from queries: "how", "to", "the", etc.
Search Options:
Option
Default
Description
--limit N
25
Maximum number of results to return
--no-limit
-
Return all matching results (no limit)
--min-score N
-
Only return results with relevance score >= N
--fast
-
Index-only search (skip content grep)
--separate
-
Show index matches and content matches separately
--no-context
-
Hide grep context lines in content matches
--clear-cache
-
Rebuild search cache before operation
--category
-
Filter results by category
--tags
-
Filter results by tags
--json
-
Output results as JSON
--verbose
-
Show relevance scores for debugging
When results are truncated, output shows "showing X of Y total" to indicate more results are available.
Hybrid Search (Default Behavior)
By default, search performs both index and content search:
Index search - Fast keyword matching against document metadata (title, description, keywords, tags)
Content search - Grep through actual file contents for matches not in index
Result markers:
[SUBSECTION] - Match found in a specific document section (click to extract)
[CONTENT] - Match found in file content (not in metadata index)
No marker - Standard index match
Performance options:
Use --fast for index-only search (faster, but may miss content-only matches)
Use --separate to display index and content matches in separate sections
Search Scoring
Results are ranked by a multi-factor scoring system:
Title matches (highest weight, 14 points)
Description matches (10 points)
Keyword field matches (8-10 points, with variant detection)
Tag matches (5-7 points)
Coverage multipliers (1.5-2.0x for multi-term matches)
Domain weighting (Claude Code docs prioritized over general Anthropic docs)
Use --verbose to see scores for debugging relevance issues.
find_docs.py Command Reference
Command
Purpose
Subsection Support
resolve <doc_id>
Resolve doc_id to file path
❌ No
content <doc_id>
Get document content
✅ Yes (--section)
search <keywords>
Keyword search
❌ No
query "<text>"
Natural language search
❌ No
category <name>
List docs in category
❌ No
tag <name>
List docs with tag
❌ No
related <doc_id>
Find related documents
❌ No
Related Documents Example:
# Find documents related to skills documentation
python .claude/skills/docs-management/scripts/core/find_docs.py related code-claude-com-docs-en-skills
# Limit results
python .claude/skills/docs-management/scripts/core/find_docs.py --limit 5 related code-claude-com-docs-en-skills
Related documents are scored by shared tags (3x weight) and shared keywords (2x weight).
Subsection Extraction (Token-Optimized)
To extract a specific section from a document (60-90% token savings):
# List all sections in a document
python scripts/core/get_subsection_content.py code-claude-com-docs-en-skills --list-sections
# Output shows hierarchical structure with heading levels# # Agent Skills# ## Create your first Skill# ## How Skills work# ### Where Skills live
Fuzzy Section Matching:
Section names support fuzzy matching - partial or word-overlap matches work:
Fetch documentation from official sources and store in canonical storage. Features: sitemap/docs map parsing, HTML→Markdown conversion, direct .md URL fetching (30-40% token savings), automatic metadata tracking, domain-based folder organization.
Detect new and removed documentation pages from sitemaps, and detect content changes via hash comparison. Features: new/removed page detection, content hash comparison, automatic stale marking, change reporting and audit logs.
Discover and resolve documentation references using doc_id, keywords, or natural language queries. Features: doc_id resolution, keyword-based search, natural language query processing, subsection discovery and extraction, category and tag filtering, alias resolution.
5. Index Management and Maintenance
Maintain index metadata, keywords, tags, and rebuild index from filesystem. Scripts: manage_index.py, rebuild_index.py, generate_report.py, verify_index.py.
6. Age-Based Cleanup
Remove documentation that has aged out based on published_at dates. Anthropic sources (engineering, news, research) have a max_age_days threshold configured in references/sources.json. Articles older than this threshold are skipped during scraping; this cleanup removes any previously-scraped articles that have since aged out.
# Clean up aged-out Anthropic articles (dry-run by default)# Reads max_age_days from sources.json automatically
python scripts/maintenance/cleanup_old_anthropic_docs.py
# Execute cleanup (actually delete files)
python scripts/maintenance/cleanup_old_anthropic_docs.py --execute
# Override with custom age threshold if needed
python scripts/maintenance/cleanup_old_anthropic_docs.py --max-age 90 --execute
This should be run after scraping and validation to ensure the canonical directory stays clean.
Workflows
Common maintenance and operational workflows for documentation management:
Adding New Documentation Source - Onboarding new docs from sitemaps or docs maps
Scraping Multiple Sources - Validation checkpoints to prevent wasted time/tokens
Token-Optimized Subsection Retrieval - Workflow for extracting subsections instead of full documents
Version bumps: Patch for doc refresh, Minor for new sources/features, Major for breaking changes
Testing: Run manage_index.py verify and test search before pushing
Development Mode
When developing this plugin locally, you may want changes to go to your dev repo instead of the installed plugin location. This skill supports explicit dev/prod mode separation via environment variable.
How It Works
By default, scripts write to wherever the plugin is installed (typically ~/.claude/plugins/marketplaces/...). When OFFICIAL_DOCS_DEV_ROOT is set to a valid skill directory, all paths resolve to that location instead.
Enabling Dev Mode
One-time setup:
# Navigate to your dev repo skill directorycd /path/to/your/claude-code-plugins/plugins/claude-ecosystem/skills/docs-management
# Generate shell commands for your shell
python scripts/setup/enable_dev_mode.py
When you run any major script (scrape, refresh, rebuild), a mode banner will display:
Dev mode:
[DEV MODE] Using development skill directory:
D:\repos\gh\claude-code-plugins\plugins\claude-ecosystem\skills\docs-management
Set via: OFFICIAL_DOCS_DEV_ROOT
Canonical dir: D:\...\canonical
Prod mode:
[PROD MODE] Using installed skill directory
(Set OFFICIAL_DOCS_DEV_ROOT to enable dev mode)
Development Workflow
Set OFFICIAL_DOCS_DEV_ROOT in your terminal
Run scripts - output goes to dev repo
Track changes: git diff canonical/
Commit and push when ready
Disabling Dev Mode
PowerShell:
Remove-Item Env:OFFICIAL_DOCS_DEV_ROOT
Bash/Zsh:
unset OFFICIAL_DOCS_DEV_ROOT
Related Documentation
index.yaml (in canonical storage) - Complete registry of docs and extracts
