Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Search for libraries in Context7. Use when the user mentions a library by name and you need to find its exact Context7 ID before fetching documentation.
.claude/skills/valtterimelkko-context7-search/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-09 | ✗→✓ | ▲ Improved | -32% | 0% |
| case-10 | ✗→✓ | ▲ Improved | 3% | 0% |
| case-14 | ✗→✓ | ▲ Improved | 6% | 0% |
| case-20 | ✗→✓ | ▲ Improved | -18% | 0% |
| case-21 | ✗→✓ | ▲ Improved | -3% | 0% |
Search for libraries in the Context7 documentation database. This helps identify the correct owner/repo format needed for fetching documentation.
Use this skill when:
Skip this skill when you already know the exact library (e.g., vercel/next.js). Use context7-docs directly instead.
Make CONTEXT7_API_KEY available either in the current environment or in a shell startup file such as ~/.bashrc. The bundled scripts can read either form.
bashpython3 ./scripts/resolve_library.py --query "LIBRARY_NAME"
bashpython3 ./scripts/resolve_library.py --query "LIBRARY_NAME" --limit 5
| Parameter | Required | Default | Description | |-----------|----------|---------|-------------| | --query | Yes | - | Library name to search (fuzzy matching) | | --limit | No | 10 | Maximum number of results |
bash# Search for Next.js python3 ./scripts/resolve_library.py --query "next.js" # Search for Prisma python3 ./scripts/resolve_library.py --query "prisma" --limit 5 # Search for MongoDB driver python3 ./scripts/resolve_library.py --query "mongodb"
json{ "success": true, "data": { "query": "next.js", "count": 3, "matches": [ { "id": "/vercel/next.js", "title": "Next.js", "description": "React framework for full-stack web applications", "stars": 131745, "trustScore": 10, "versions": ["v15.1.8", "v14.3.0", "v13.5.11"] } ] } }
After getting results, select the best library based on:
For frequently used libraries, you can skip search and use directly:
| Library | ID | |---------|-------------| | Next.js | vercel/next.js | | React | facebook/react | | Prisma | prisma/prisma | | Supabase | supabase/supabase | | FastAPI | fastapi/fastapi | | Express | expressjs/express | | Django | django/django |
1. User asks about a library (e.g., "How do I use Prisma?")
|
2. Run context7-search to find library ID
|
3. Select best match from results (e.g., prisma/prisma)
|
4. Use context7-docs skill with the library IDNo matches found:
API error:
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-05 | fail→fail | 4,138 | 4,285 | +4% | 1 | 1 | 0% | 598 | 1,026 | +72% | 0 | 0 | — |
case-06 | pass→pass | 2,625 | 2,027 | -23% | 1 | 1 | 0% | 441 | 1,152 | +161% | 0 | 0 | — |
case-07 | pass→pass | 3,449 | 2,312 | -33% | 1 | 1 | 0% | 499 | 1,155 | +131% | 0 | 0 | — |
case-01 | fail→fail | 2,199 | 14,823 | +574% | 1 | 1 | 0% | 231 | 1,103 | +377% | 0 | 0 | — |
case-02 | fail→fail | 4,878 | 10,758 | +121% | 1 | 1 | 0% | 214 | 1,332 | +522% | 0 | 0 | — |
case-03 | fail→fail | 9,184 | 9,509 | +4% | 1 | 1 | 0% | 766 | 1,106 | +44% | 0 | 0 | — |
case-04 | fail→fail | 5,603 | 8,552 | +53% | 1 | 1 | 0% | 431 | 1,665 | +286% | 0 | 0 | — |
case-08 | pass→pass | 3,559 | 1,967 | -45% | 1 | 1 | 0% | 525 | 1,185 | +126% | 0 | 0 | — |
case-09 | fail→pass | 10,021 | 1,715 | -83% | 1 | 1 | 0% | 1,631 | 1,105 | -32% | 0 | 0 | — |
case-10 | fail→pass | 7,636 | 2,021 | -74% | 1 | 1 | 0% | 1,153 | 1,191 | +3% | 0 | 0 | — |
case-11 | pass→pass | 5,132 | 3,600 | -30% | 1 | 1 | 0% | 768 | 1,356 | +77% | 0 | 0 | — |
case-12 | pass→pass | 3,859 | 1,466 | -62% | 1 | 1 | 0% | 633 | 1,058 | +67% | 0 | 0 | — |
case-13 | pass→pass | 14,004 | 9,471 | -32% | 1 | 1 | 0% | 822 | 1,111 | +35% | 0 | 0 | — |
case-14 | fail→pass | 6,862 | 1,546 | -77% | 1 | 1 | 0% | 1,040 | 1,100 | +6% | 0 | 0 | — |
case-15 | pass→pass | 4,510 | 1,538 | -66% | 1 | 1 | 0% | 628 | 1,049 | +67% | 0 | 0 | — |
case-16 | pass→pass | 5,225 | 2,035 | -61% | 1 | 1 | 0% | 634 | 1,115 | +76% | 0 | 0 | — |
case-17 | fail→fail | 9,927 | 5,260 | -47% | 1 | 1 | 0% | 1,728 | 1,219 | -29% | 0 | 0 | — |
case-18 | fail→fail | 15,039 | 6,248 | -58% | 1 | 1 | 0% | 2,183 | 1,188 | -46% | 0 | 0 | — |
case-19 | pass→pass | 4,169 | 2,458 | -41% | 1 | 1 | 0% | 739 | 1,280 | +73% | 0 | 0 | — |
case-20 | fail→pass | 19,081 | 1,786 | -91% | 1 | 1 | 0% | 1,336 | 1,093 | -18% | 0 | 0 | — |
case-21 | fail→pass | 7,829 | 2,777 | -65% | 1 | 1 | 0% | 1,423 | 1,376 | -3% | 0 | 0 | — |
case-22 | pass→pass | 9,926 | 1,375 | -86% | 1 | 1 | 0% | 1,589 | 1,001 | -37% | 0 | 0 | — |
case-23 | fail→fail | 7,563 | 4,842 | -36% | 1 | 1 | 0% | 1,266 | 1,004 | -21% | 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. 23 cases were attempted, and 15 counted toward the lift figure. The other 8 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 +22 percentage points is the difference between those two pass rates over the 15 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.