Not in the top 1,000 by GitHub stars

costiash/claude-code-docs

Always-current official Claude docs inside Claude Code. A 3 MB metadata plugin: live on-demand fetching from Anthropic's servers, zero content redistributed. Auto-discovery skill, instant search, interactive HTML courses, and doc changelog reports.

History

Review weekly changes to SKILL.md before updating. Dates mark the weeks in which we observed each version; checks apply to the recorded commits.

View weekly changes

plugin/skills/claude-docs/SKILL.md

Week of Nov 2, 2026 · +0 / −0 lines

Before Not checked yet → After Network or broad shell

f217584338a4 → 5a9de91bc2bd

Diff text is unavailable under the recorded licenses. Line counts and the GitHub link remain available.

Compare on GitHub

Week of Oct 26, 2026 · Line counts unavailable

Before Not checked yet → After Not checked yet

First observation → f217584338a4

No earlier version was observed for comparison.

Compare on GitHub

plugin/skills/claude-docs-changelog/SKILL.md

Week of Nov 2, 2026 · +0 / −0 lines

Before Not checked yet → After Network or broad shell

f217584338a4 → 5a9de91bc2bd

Diff text is unavailable under the recorded licenses. Line counts and the GitHub link remain available.

Compare on GitHub

Week of Oct 26, 2026 · Line counts unavailable

Before Not checked yet → After Not checked yet

First observation → f217584338a4

No earlier version was observed for comparison.

Compare on GitHub

plugin/skills/claude-docs-course/SKILL.md

Week of Nov 2, 2026 · +0 / −0 lines

Before Not checked yet → After Network or broad shell

f217584338a4 → 5a9de91bc2bd

Diff text is unavailable under the recorded licenses. Line counts and the GitHub link remain available.

Compare on GitHub

Week of Oct 26, 2026 · Line counts unavailable

Before Not checked yet → After Not checked yet

First observation → f217584338a4

No earlier version was observed for comparison.

Compare on GitHub

plugin/skills/claude-docs-validate/SKILL.md

Week of Nov 2, 2026 · +0 / −0 lines

Before Not checked yet → After Network or broad shell

f217584338a4 → 5a9de91bc2bd

Diff text is unavailable under the recorded licenses. Line counts and the GitHub link remain available.

Compare on GitHub

Week of Oct 26, 2026 · Line counts unavailable

Before Not checked yet → After Not checked yet

First observation → f217584338a4

No earlier version was observed for comparison.

Compare on GitHub

plugin/skills/docs/SKILL.md

Week of Nov 2, 2026 · +0 / −0 lines

Before Not checked yet → After Network or broad shell

f217584338a4 → 5a9de91bc2bd

Diff text is unavailable under the recorded licenses. Line counts and the GitHub link remain available.

Compare on GitHub

Week of Oct 26, 2026 · Line counts unavailable

Before Not checked yet → After Not checked yet

First observation → f217584338a4

No earlier version was observed for comparison.

Compare on GitHub

View on GitHubAdd to my skills

Verified commit 5a9de91bc2bd77e09faebbd7ff421e2cede32215

git clone --filter=blob:none --no-checkout https://github.com/costiash/claude-code-docs.git .hotskills/costiash-claude-code-docs && git -C .hotskills/costiash-claude-code-docs checkout --detach 5a9de91bc2bd77e09faebbd7ff421e2cede32215 # Copy each selected skill directory from .hotskills/costiash-claude-code-docs into .claude/skills/<skill-name>/
56GitHub stars
0stars in 7 days
5skills in this repository
133commits in 30 days

Heat

1.8
  • Star growth, last 7 daysNot available
  • Star growth, last 24 hoursNot available
  • Rank movement, last 7 daysNot available
  • Commits, last 30 daysNot available
  • Accounts asking about it through MCP, last 7 daysNot available
  • Hot votes this weekNot available

Heat never changes the order by stars, and sponsorship never touches it. How heat works

Stars over time

Oct 25, 2026Oct 30, 202656

30-day change not yet available: not enough history.

Individual skills in this repository

How it works

claude-docs

When your agent uses it

Search and read locally-stored Claude documentation covering Claude Code CLI, Claude API (Messages, tool use, vision, streaming, batch), Agent SDK (Python and TypeScript), prompt engineering, and all Anthropic platform docs. Use this skill whenever the user asks about Claude Code features (hooks, MCP servers, skills, plugins, settings, permissions, keybindings, sub-agents), the Anthropic API or any of its SDKs (Python, TypeScript, Go, Java), the Agent SDK (sessions, hooks, custom tools, MCP), model capabilities (context windows, extended thinking, pricing, rate limits, vision), prompt engineering best practices, or troubleshooting any Claude-related error. This skill provides instant access to official documentation files without web searches — always prefer it over web lookups for Claude and Anthropic topics.

What it loads into context

Frontmatter

192 tokens

Always available to the agent

SKILL.md

1,766 tokens

When the skill is activated

Reference documents

2,393 tokens

Loaded on request

Token counts are estimates. Scripts run separately; they are not counted as reference context.

The playbook

  1. When to Use This Skill
  2. Search Strategy
  3. Reading a doc (cache-miss rule)
  4. Synthesis Rules
  5. URL Generation
  6. Common Synonyms
  7. No Results
  8. Post-Response Course Prompt
  9. Reference Files

Scripts executed separately

  • ../../hooks/sync-docs.sh
  • scripts/content-search.sh
  • scripts/fuzzy-search.sh

Files

  • ../../../.claude-plugin/marketplace.jsonasset659 bytes
  • ../../.claude-plugin/plugin.jsonasset633 bytes
  • ../../hooks/hooks.jsonasset242 bytes
  • ../../hooks/sync-docs.shruns22,948 bytes
  • SKILL.mdcore7,125 bytes
  • SKILL.mdcore7,125 bytes
  • SKILL.mdcore7,125 bytes
  • SKILL.mdcore7,125 bytes
  • SKILL.mdcore7,125 bytes
  • SKILL.mdcore7,125 bytes
  • SKILL.mdcore7,125 bytes
  • SKILL.mdcore7,125 bytes
  • SKILL.mdcore7,125 bytes
  • SKILL.mdcore7,125 bytes
  • SKILL.mdcore7,125 bytes
  • SKILL.mdcore7,125 bytes
  • SKILL.mdcore7,125 bytes
  • SKILL.mdcore7,125 bytes
  • examples/cross-context.mddoc2,435 bytes
  • examples/direct-lookup.mddoc1,864 bytes
  • examples/semantic-search.mddoc2,046 bytes
  • manifest-reference.mddoc2,638 bytes
  • scripts/content-search.shruns3,579 bytes
  • scripts/fuzzy-search.shruns3,283 bytes

No excerpt is shown for this license. Read the source on GitHub.

View the complete source on GitHub

Skills included

  • claude-docs
  • claude-docs-changelog
  • claude-docs-course
  • claude-docs-validate
  • docs

Repository facts

License
No license detected
Created
Nov 3, 2025
Last pushed
Nov 7, 2026
Forks
5
Open issues
0
Project activity
Active

Before you install

Network or broad shell

Automated signals, not a guarantee

Checked Nov 7, 2026

Not fully analyzed: some files were too long, unreadable or missing. This never lowers the level.

Show 31 notes
  • plugin/skills/claude-docs/SKILL.md:1 Analysis incomplete: Referenced executable target unavailable; not analyzed
  • plugin/skills/claude-docs/SKILL.md:1 Analysis incomplete: Referenced executable target unavailable; not analyzed
  • plugin/skills/claude-docs/SKILL.md:1 Analysis incomplete: Referenced executable target unavailable; not analyzed
  • plugin/skills/claude-docs/SKILL.md:1 Analysis incomplete: Referenced executable target unavailable; not analyzed
  • plugin/skills/claude-docs/SKILL.md:1 Analysis incomplete: Referenced executable target unavailable; not analyzed
  • plugin/skills/claude-docs/SKILL.md:1 Analysis incomplete: Referenced executable target unavailable; not analyzed
  • plugin/skills/claude-docs/SKILL.md:1 Analysis incomplete: Referenced executable target unavailable; not analyzed
  • plugin/skills/claude-docs/SKILL.md:1 Analysis incomplete: Referenced executable target unavailable; not analyzed
  • plugin/skills/claude-docs/SKILL.md:1 Analysis incomplete: Referenced executable target unavailable; not analyzed
  • plugin/skills/claude-docs/SKILL.md:1 Analysis incomplete: Referenced executable target unavailable; not analyzed
  • plugin/skills/claude-docs/SKILL.md:1 Analysis incomplete: Referenced executable target unavailable; not analyzed
  • plugin/skills/claude-docs/SKILL.md:1 Analysis incomplete: Referenced executable target unavailable; not analyzed
  • plugin/skills/claude-docs/SKILL.md:1 Analysis incomplete: Referenced executable target unavailable; not analyzed
  • plugin/skills/claude-docs-changelog/SKILL.md:1 Analysis incomplete: Referenced executable target unavailable; not analyzed
  • plugin/skills/claude-docs-changelog/SKILL.md:1 Analysis incomplete: Referenced executable target unavailable; not analyzed
  • plugin/skills/claude-docs-changelog/SKILL.md:1 Analysis incomplete: Referenced executable target unavailable; not analyzed
  • plugin/skills/claude-docs-course/SKILL.md:1 Analysis incomplete: Referenced executable target unavailable; not analyzed
  • plugin/skills/claude-docs-course/SKILL.md:1 Analysis incomplete: Referenced executable target unavailable; not analyzed
  • plugin/skills/claude-docs-course/SKILL.md:1 Analysis incomplete: Referenced executable target unavailable; not analyzed
  • plugin/skills/claude-docs-course/SKILL.md:1 Analysis incomplete: Referenced executable target unavailable; not analyzed
Show all 79 signals by file

.claude-plugin/marketplace.json

  • Network accessline 11
    "name": "claude-docs",       "source": "./plugin",       "homepage": "https://costiash.github.io/claude-code-docs/",
  • Network accessline 12
    "source": "./plugin",       "homepage": "https://costiash.github.io/claude-code-docs/",       "category": "documentation",
  • Network accessline 13
    "homepage": "https://costiash.github.io/claude-code-docs/",

plugin/.claude-plugin/plugin.json

  • Network accessline 6
    "name": "costiash"   },   "homepage": "https://costiash.github.io/claude-code-docs/",
  • Network accessline 7
    },   "homepage": "https://costiash.github.io/claude-code-docs/",   "repository": "https://github.com/costiash/claude-code-docs",
  • Network accessline 8
    "homepage": "https://costiash.github.io/claude-code-docs/",
  • Network accessline 9
    "repository": "https://github.com/costiash/claude-code-docs",

plugin/hooks/sync-docs.sh

  • Executable scriptline 1
    #!/bin/bash
  • Network accessline 16
    REPO_URL="$CLAUDE_DOCS_REPO_URL" else     REPO_URL="https://github.com/costiash/claude-code-docs.git"
  • Network accessline 17
    else     REPO_URL="https://github.com/costiash/claude-code-docs.git" fi
  • Network accessline 18
    REPO_URL="https://github.com/costiash/claude-code-docs.git" fi FETCH="$DOCS_DIR/plugin/scripts/fetch-docs.sh"
  • Dangerous shell operationline 172
    rm -rf "$DOCS_DIR/$sub/.sync.lock" 2>/dev/null || true
  • Dangerous shell operationline 196
    rmdir "$DOCS_DIR.heal.lock" 2>/dev/null || rm -rf "$DOCS_DIR.heal.lock" 2>/dev/null
  • Dangerous shell operationline 209
    run_with_timeout "$(cap 10)" rm -rf "$d" >/dev/null 2>&1 || true
  • Network accessline 222
    exit 0     fi     if run_with_timeout "$(cap 30)" git clone --depth 1 "$REPO_URL" "$DOCS_DIR" >/dev/null 2>&1; then
  • Network accessline 223
    fi     if run_with_timeout "$(cap 30)" git clone --depth 1 "$REPO_URL" "$DOCS_DIR" >/dev/null 2>&1; then         prune_swap_orphans
  • Network accessline 224
    if run_with_timeout "$(cap 30)" git clone --depth 1 "$REPO_URL" "$DOCS_DIR" >/dev/null 2>&1; then
  • Network accessline 227
    output_context "Claude documentation installed ($(doc_count) pages indexed). Pages download to the local cache in the background; /docs works immediately (missing pages fetch on demand)."     else         output_context "Failed to clone Cla
  • Network accessline 228
    else         output_context "Failed to clone Claude documentation. Run: git clone $REPO_URL $DOCS_DIR"     fi
  • Network accessline 229
    output_context "Failed to clone Claude documentation. Run: git clone $REPO_URL $DOCS_DIR"
  • Dangerous shell operationline 267
    #   clone fails                -> rm NEW (no user data yet), keep DOCS_DIR
  • Dangerous shell operationline 272
    #   success                    -> rm OLD (capped); orphans: none
  • Dangerous shell operationline 278
    # otherwise leapfrog each other's swaps and rm -rf a freshly-healed
  • Dangerous shell operationline 287
    rmdir "$HEAL_LOCK" 2>/dev/null || rm -rf "$HEAL_LOCK" 2>/dev/null
  • Dangerous shell operationline 293
    trap 'rmdir "$HEAL_LOCK" 2>/dev/null || rm -rf "$HEAL_LOCK" 2>/dev/null' EXIT
  • Dangerous shell operationline 301
    # would end in the clone-failed branch's rm -rf of NEW_DIR deleting it
  • Dangerous shell operationline 309
    run_with_timeout "$(cap 10)" rm -rf "$d" >/dev/null 2>&1 || true
  • Network accessline 325
    if run_with_timeout "$(cap 30)" git clone --depth 1 "$REPO_URL" "$NEW_DIR" >/dev/null 2>&1 \
  • Dangerous shell operationline 333
    # be rm -rf'd on the success path — the one asymmetry against the
  • Dangerous shell operationline 354
    # rm -rf left survivors and mv nested the fresh clone inside them.
  • Dangerous shell operationline 375
    # rm -rf the lot. Defer instead: both temp dirs outlive this process
  • Dangerous shell operationline 399
    run_with_timeout "$(cap 10)" rm -rf "$OLD_DIR" >/dev/null 2>&1 || true
  • Dangerous shell operationline 406
    run_with_timeout "$(cap 10)" rm -rf "$NEW_DIR" >/dev/null 2>&1 || true

plugin/skills/claude-docs-changelog/SKILL.md

  • Network accessline 125
    <link href="https://fonts.googleapis.com/css2?family=Instrument+Serif:ital@0;1&family=Outfit:wght@300;400;500;600;700&family=JetBrains+Mono:wght@400;500&display=swap" rel="stylesheet">

plugin/skills/claude-docs-changelog/examples/changelog-report.md

  • Network accessline 40
    2. Structure: header → stats bar → highlights → CLI section (5 cards) → SDK section (grouped cards) 3. Each card's source link is the entry's `.url` verbatim (e.g.    `https://code.claude.com/docs/en/agent-sdk/tool-search`) — never reconstr
  • Network accessline 41
    3. Each card's source link is the entry's `.url` verbatim (e.g.    `https://code.claude.com/docs/en/agent-sdk/tool-search`) — never reconstructed from the filename 4. Each card gets a "Create Course" button with `data-topic` extracted from 
  • Network accessline 42
    `https://code.claude.com/docs/en/agent-sdk/tool-search`) — never reconstructed from the filename

plugin/skills/claude-docs-course/examples/course-from-docs.md

  • Network accessline 62
    - Multiple-choice quizzes testing practical hook configuration - Glossary tooltips on terms: matchers, IIFE, ARIA, stdin, SSE, MCP - Source links from the manifest, e.g. [Hooks reference](https://code.claude.com/docs/en/hooks),
  • Network accessline 63
    - Glossary tooltips on terms: matchers, IIFE, ARIA, stdin, SSE, MCP - Source links from the manifest, e.g. [Hooks reference](https://code.claude.com/docs/en/hooks),   [Automate actions with hooks](https://code.claude.com/docs/en/hooks-guide
  • Network accessline 64
    - Source links from the manifest, e.g. [Hooks reference](https://code.claude.com/docs/en/hooks),
  • Network accessline 65
    [Automate actions with hooks](https://code.claude.com/docs/en/hooks-guide),
  • Network accessline 66
    [Agent SDK hooks](https://code.claude.com/docs/en/agent-sdk/hooks)

plugin/skills/claude-docs-course/references/design-system.md

  • Network accessline 111
    <link rel="preconnect" href="https://fonts.googleapis.com">
  • Network accessline 112
    <link rel="preconnect" href="https://fonts.gstatic.com" crossorigin>
  • Network accessline 113
    <link href="https://fonts.googleapis.com/css2?family=Instrument+Serif:ital@0;1&family=Outfit:wght@300;400;500;600;700&family=JetBrains+Mono:wght@400;500;600&display=swap" rel="stylesheet">
  • Network accessline 521
    z-index: 9999;   opacity: 0.035;   background-image: url("data:image/svg+xml,%3Csvg viewBox='0 0 256 256' xmlns='http://www.w3.org/2000/svg'%3E%3Cfilter id='noise'%3E%3CfeTurbulence type='fractalNoise' baseFrequency='0.9' numOctaves='4' sti
  • Network accessline 522
    opacity: 0.035;   background-image: url("data:image/svg+xml,%3Csvg viewBox='0 0 256 256' xmlns='http://www.w3.org/2000/svg'%3E%3Cfilter id='noise'%3E%3CfeTurbulence type='fractalNoise' baseFrequency='0.9' numOctaves='4' stitchTiles='stitch'
  • Network accessline 523
    background-image: url("data:image/svg+xml,%3Csvg viewBox='0 0 256 256' xmlns='http://www.w3.org/2000/svg'%3E%3Cfilter id='noise'%3E%3CfeTurbulence type='fractalNoise' baseFrequency='0.9' numOctaves='4' stitchTiles='stitch'/%3E%3C/filter%3E%

plugin/skills/claude-docs-course/references/interactive-elements.md

  • Network accessline 564
    <div class="bug-line bug-target" data-line="3" onclick="checkBugLine(this, true)">       <span class="line-num">3</span>       <code>    fetch(url).then(r => r.json()).then(data => sendResponse(data));</code>
  • Network accessline 565
    <span class="line-num">3</span>       <code>    fetch(url).then(r => r.json()).then(data => sendResponse(data));</code>     </div>
  • Network accessline 566
    <code>    fetch(url).then(r => r.json()).then(data => sendResponse(data));</code>

plugin/skills/claude-docs-validate/SKILL.md

  • Network accessline 58
    These read URLs directly from the manifest. Report the summary (reachable / broken / timed out). For persistent broken URLs, the upstream page may have moved — report at https://github.com/costiash/claude-code-docs/issues.
  • Network accessline 60
    https://github.com/costiash/claude-code-docs/issues.

plugin/skills/claude-docs-validate/examples/validate-docs.md

  • Network accessline 46
    pages may have moved. The SessionStart hook refreshes the metadata automatically; a manual refresh is `cd ~/.claude-code-docs && git fetch origin main && git reset --hard origin/main`. Report persistent issues at https://github.com/costiash
  • Network accessline 48
    Report persistent issues at https://github.com/costiash/claude-code-docs/issues."

plugin/skills/claude-docs-validate/scripts/validate-paths.sh

  • Executable scriptline 1
    #!/usr/bin/env bash
  • Network accessline 50
    status=$(curl -sI --proto '=https' --max-time "${TIMEOUT:-10}" -o /dev/null -w "%{http_code}" "$url" 2>/dev/null || echo "000")

plugin/skills/claude-docs/SKILL.md

  • Network accessline 132
    Example: `claude-code__hooks.md` → `https://code.claude.com/docs/en/hooks`.

plugin/skills/claude-docs/examples/cross-context.md

  • Network accessline 36
    `~/.claude-code-docs/plugin/scripts/fetch-docs.sh get "claude-code__skills.md"` first), synthesize, present with the manifest's source link ([Extend Claude with skills](https://code.claude.com/docs/en/skills)).
  • Network accessline 38
    ([Extend Claude with skills](https://code.claude.com/docs/en/skills)).
  • Network accessline 53
    for the endpoint parameters 3. Present the Python SDK example 4. Note: "TypeScript equivalent: [TypeScript SDK](https://platform.claude.com/docs/en/cli-sdks-libraries/sdks/typescript)"
  • Network accessline 55
    4. Note: "TypeScript equivalent: [TypeScript SDK](https://platform.claude.com/docs/en/cli-sdks-libraries/sdks/typescript)"
  • Network accessline 57
    Sources: - [Python SDK](https://platform.claude.com/docs/en/cli-sdks-libraries/sdks/python)
  • Network accessline 58
    - [Python SDK](https://platform.claude.com/docs/en/cli-sdks-libraries/sdks/python)
  • Network accessline 59
    - [Create a Message](https://platform.claude.com/docs/en/api/messages/create)

plugin/skills/claude-docs/examples/direct-lookup.md

  • Network accessline 35
    For Agent SDK hooks, see also the [Agent SDK hooks guide](https://code.claude.com/docs/en/agent-sdk/hooks).
  • Network accessline 37
    Sources: - [Hooks reference](https://code.claude.com/docs/en/hooks)
  • Network accessline 38
    - [Hooks reference](https://code.claude.com/docs/en/hooks)
  • Network accessline 39
    - [Automate actions with hooks](https://code.claude.com/docs/en/hooks-guide)
  • Network accessline 40
    - [Intercept and control agent behavior with hooks](https://code.claude.com/docs/en/agent-sdk/hooks)"

plugin/skills/claude-docs/examples/semantic-search.md

  • Network accessline 39
    Sources: - [Thinking overview](https://platform.claude.com/docs/en/build-with-claude/thinking)
  • Network accessline 40
    - [Thinking overview](https://platform.claude.com/docs/en/build-with-claude/thinking)
  • Network accessline 41
    - [Extended thinking (legacy)](https://platform.claude.com/docs/en/build-with-claude/extended-thinking)
  • Network accessline 42
    - [Steering and cost control](https://platform.claude.com/docs/en/build-with-claude/thinking-steering-and-cost)"

plugin/skills/claude-docs/manifest-reference.md

  • Network accessline 53
    Example: `claude-code__hooks.md` → `https://code.claude.com/docs/en/hooks`.

plugin/skills/claude-docs/scripts/content-search.sh

  • Executable scriptline 1
    #!/usr/bin/env bash
  • Dangerous shell operationline 76
    tmp=$(mktemp); trap 'rm -f "$tmp"' EXIT

plugin/skills/claude-docs/scripts/fuzzy-search.sh

  • Executable scriptline 1
    #!/usr/bin/env bash

plugin/skills/docs/SKILL.md

  • Dangerous shell operationline 67
    → Optionally clean up the clone and cache: `rm -rf ~/.claude-code-docs` (this also removes `~/.claude-code-docs/cache/` and any generated courses).

How the signals fit together

Facts from a static read of the files at the pinned commit. They describe what the code does, not why.

Some files could not be read; the facts below still hold.

Signal confidence: 72 facts

and 22 more

Works with

Agent compatibility, from the files in this repository
AgentStatusEvidenceInstall to
Claude CodeNativeManifest .claude-plugin/marketplace.json.claude/skills/<skill-name>/
CodexCompatibleStandard SKILL.md plugin/skills/claude-docs/SKILL.md.agents/skills/<skill-name>/
CursorCompatibleStandard SKILL.md plugin/skills/claude-docs/SKILL.md.cursor/skills/<skill-name>/
Gemini CLICompatibleStandard SKILL.md plugin/skills/claude-docs/SKILL.md.gemini/skills/<skill-name>/
GitHub CopilotCompatibleStandard SKILL.md plugin/skills/claude-docs/SKILL.md.github/skills/<skill-name>/
OpenCodeCompatibleStandard SKILL.md plugin/skills/claude-docs/SKILL.md.opencode/skills/<skill-name>/
  • Native The repository ships a manifest, folder or frontmatter field for this agent.
  • Compatible A standard SKILL.md that this agent can read.
  • Unknown No evidence either way.

Derived automatically from paths, manifests and frontmatter. Not tested on each agent.