Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Read and search GitHub repository documentation via gitmcp.io MCP service. **WHEN TO USE:** - User provides a GitHub URL - User mentions a specific repo in owner/repo format - User asks "what does this repo do?", "read the docs for X repo", or similar - User wants to search code or docs within a repo
.claude/skills/aiskillstore-read-github/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-07 | ✗→✓ | ▲ Improved | -42% | 0% |
| case-08 | ✗→✓ | ▲ Improved | -37% | 0% |
| case-09 | ✗→✓ | ▲ Improved | -71% | 0% |
| case-10 | ✗→✓ | ▲ Improved | -26% | 0% |
| case-11 | ✗→✓ | ▲ Improved | -71% | 0% |
Access GitHub repository documentation and code via the gitmcp.io MCP service.
Convert GitHub URLs to gitmcp.io:
github.com/owner/repo → gitmcp.io/owner/repohttps://github.com/karpathy/llm-council → https://gitmcp.io/karpathy/llm-councilThe scripts/gitmcp.py script provides CLI access to repository docs.
bashpython3 scripts/gitmcp.py list-tools owner/repo
Retrieves the full documentation file (README, docs, etc.):
bashpython3 scripts/gitmcp.py fetch-docs owner/repo
Semantic search within repository documentation:
bashpython3 scripts/gitmcp.py search-docs owner/repo "query"
Search code using GitHub Search API (exact match):
bashpython3 scripts/gitmcp.py search-code owner/repo "function_name"
Fetch content from URLs mentioned in documentation:
bashpython3 scripts/gitmcp.py fetch-url owner/repo "https://example.com/doc"
Call any MCP tool directly:
bashpython3 scripts/gitmcp.py call owner/repo tool_name '{"arg": "value"}'
Tool names are dynamically prefixed with the repo name (underscored):
karpathy/llm-council → fetch_llm_council_documentationfacebook/react → fetch_react_documentationmy-org/my-repo → fetch_my_repo_documentationFor any repository, these tools are available:
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-04 | pass→pass | 14,318 | 22,128 | +55% | 1 | 1 | 0% | 1,591 | 1,859 | +17% | 0 | 0 | — |
case-01 | fail→fail | 16,054 | 10,636 | -34% | 1 | 1 | 0% | 303 | 785 | +159% | 0 | 0 | — |
case-02 | fail→fail | 17,993 | 10,720 | -40% | 1 | 1 | 0% | 2,085 | 915 | -56% | 0 | 0 | — |
case-03 | fail→fail | 24,911 | 15,485 | -38% | 1 | 1 | 0% | 2,908 | 898 | -69% | 0 | 0 | — |
case-05 | pass→pass | 15,842 | 11,861 | -25% | 1 | 1 | 0% | 1,720 | 2,441 | +42% | 0 | 0 | — |
case-06 | pass→pass | 11,192 | 11,009 | -2% | 1 | 1 | 0% | 812 | 1,529 | +88% | 0 | 0 | — |
case-07 | fail→pass | 14,745 | 2,089 | -86% | 1 | 1 | 0% | 1,583 | 912 | -42% | 0 | 0 | — |
case-08 | fail→pass | 8,081 | 7,043 | -13% | 1 | 1 | 0% | 1,523 | 952 | -37% | 0 | 0 | — |
case-09 | fail→pass | 25,107 | 7,342 | -71% | 1 | 1 | 0% | 3,213 | 933 | -71% | 0 | 0 | — |
case-10 | fail→pass | 12,754 | 2,748 | -78% | 1 | 1 | 0% | 1,366 | 1,015 | -26% | 0 | 0 | — |
case-11 | fail→pass | 24,526 | 2,242 | -91% | 1 | 1 | 0% | 3,169 | 931 | -71% | 0 | 0 | — |
case-12 | fail→pass | 19,474 | 4,643 | -76% | 1 | 1 | 0% | 2,319 | 1,391 | -40% | 0 | 0 | — |
case-13 | fail→pass | 6,599 | 8,252 | +25% | 1 | 1 | 0% | 974 | 1,166 | +20% | 0 | 0 | — |
case-14 | fail→pass | 12,496 | 9,278 | -26% | 1 | 1 | 0% | 2,205 | 1,108 | -50% | 0 | 0 | — |
case-15 | fail→pass | 22,060 | 8,163 | -63% | 1 | 1 | 0% | 2,974 | 1,150 | -61% | 0 | 0 | — |
case-16 | fail→pass | 23,455 | 3,323 | -86% | 1 | 1 | 0% | 3,478 | 1,108 | -68% | 0 | 0 | — |
case-21 | fail→pass | 14,268 | 9,209 | -35% | 1 | 1 | 0% | 2,367 | 1,376 | -42% | 0 | 0 | — |
case-17 | fail→pass | 23,634 | 1,772 | -93% | 1 | 1 | 0% | 3,253 | 909 | -72% | 0 | 0 | — |
case-18 | fail→pass | 18,775 | 8,141 | -57% | 1 | 1 | 0% | 2,120 | 1,144 | -46% | 0 | 0 | — |
case-19 | pass→pass | 15,945 | 6,999 | -56% | 1 | 1 | 0% | 1,747 | 884 | -49% | 0 | 0 | — |
case-20 | fail→pass | 11,052 | 7,920 | -28% | 1 | 1 | 0% | 868 | 1,092 | +26% | 0 | 0 | — |
case-22 | fail→pass | 14,408 | 7,367 | -49% | 1 | 1 | 0% | 1,523 | 977 | -36% | 0 | 0 | — |
DecimalAI ran this skill against gemini-3.6-flash twice over the same eval suite — once with the skill loaded and once without — and compared the two runs case by case. 22 cases were attempted, and 19 counted toward the lift figure. The other 3 produced results that are not comparable between the two arms, so they are excluded from the headline rather than averaged into it. The headline lift of +68 percentage points is the difference between those two pass rates over the 19 comparable cases.
Without the skill loaded, the model failed this case. With it loaded, the same prompt on the same model passed. This is one improved case from the latest verified run; every case, including any that regressed, is in the table above.
Other measured skills in the registry, with their headline benchmark lift.