aios-core
Version:
Synkra AIOS: AI-Orchestrated System for Full Stack Development - Core Framework
79 lines (70 loc) • 3.27 kB
YAML
tool:
schema_version: 1.0
id: context7
type: mcp
name: Context7 Library Documentation
version: 1.0.0
description: Retrieve up-to-date documentation and code examples for any software library via Context7 AI
knowledge_strategy: declarative
capabilities:
- Resolve library names to Context7-compatible IDs
- Fetch documentation for specific libraries
- Filter documentation by topic
- Control token usage for documentation retrieval
- Support version-specific documentation
use_cases:
- library_documentation: "Get current documentation for any library"
- api_reference: "Look up API methods and usage examples"
- version_specific_docs: "Access documentation for specific library versions"
- focused_topics: "Retrieve docs filtered by specific topics (e.g., hooks, routing)"
mcp_specific:
server_command: "sse"
transport: sse
connection:
type: sse
url: "https://mcp.context7.com/sse"
health_check:
method: tool_call
command: resolve-library-id
expected_response: "Library ID resolution"
timeout_ms: 5000
tools:
- id: resolve-library-id
description: "Resolves a package/product name to a Context7-compatible library ID"
parameters:
- libraryName: "Library name to search for (e.g., 'react', 'next.js')"
notes:
- "MUST call this before get-library-docs unless user provides explicit library ID"
- "Returns library ID in format '/org/project' or '/org/project/version'"
- "Selection based on name similarity, description relevance, and trust score"
- id: get-library-docs
description: "Fetches up-to-date documentation for a library"
parameters:
- context7CompatibleLibraryID: "Exact library ID from resolve-library-id or user (e.g., '/mongodb/docs', '/vercel/next.js')"
- topic: "Optional topic to focus documentation on (e.g., 'hooks', 'routing')"
- tokens: "Maximum tokens of documentation to retrieve (default: 5000)"
notes:
- "Must use library ID from resolve-library-id first"
- "Higher token values provide more context but consume more tokens"
workflow:
typical_usage:
- step: 1
action: "Call resolve-library-id with library name"
example: "resolve-library-id({ libraryName: 'next.js' })"
- step: 2
action: "Extract library ID from response"
example: "libraryID = '/vercel/next.js'"
- step: 3
action: "Call get-library-docs with library ID"
example: "get-library-docs({ context7CompatibleLibraryID: '/vercel/next.js', topic: 'routing' })"
best_practices:
- "Always call resolve-library-id first unless user provides explicit ID"
- "Use topic parameter to focus on specific areas of documentation"
- "Adjust tokens parameter based on how much context you need"
- "Check trust score when multiple libraries match (prefer 7-10)"
- "For ambiguous queries, request clarification before proceeding"
limitations:
- "Requires active internet connection to Context7 service"
- "Documentation availability depends on Context7 coverage"
- "Token usage counts against retrieval limits"
- "SSE connection may timeout on slow networks"