search-context
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| LOG_LEVEL | No | Log level: debug, info, or error | info |
| GEMINI_API_KEY | Yes | Your Gemini API key |
Instructions
Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.
This server publishes no instructions, or was last inspected before Glama recorded them.
Capabilities
Features and capabilities supported by this server
Protocol revision2025-11-25
| Capability | Details |
|---|---|
| tools | {
"listChanged": true
} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| ask_docs_agentA | AI-powered semantic search for complex documentation questions. Best for conceptual queries, multi-part research, and understanding how systems work. How to Use:
Required: • store: Reference name from MCP Resources (e.g., "context", "Factory-AI/factory") • query: Your question in natural language Optional (use when needed): • include_chunks: true - Show document excerpts for verification (increases tokens ~3x) • top_k: 1-20 - Number of excerpts when include_chunks=true (default: 3) • response_format: "json" - Structured output instead of markdown • metadata_filter: Advanced filtering using List Filter syntax Effective Queries: ✅ "How does authentication work and why is it designed this way?" ✅ "What are the key differences between async and sync processing?" ✅ "Explain the rate limiting strategy and its tradeoffs" ❌ "authentication" (too vague) ❌ Single keywords without context Finding References: All available documentation references are registered as MCP Resources. Use the Resources list to see what's searchable. Each top-level directory from ain3sh/docs is its own reference. |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
No prompts | |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
No resources | |
TDQS
Scored across 1 tool
With only one tool, there is no possibility of misselection or overlap with sibling tools. Its purpose (semantic documentation Q&A) is stated clearly and unambiguously in the description.
The single name ask_docs_agent is readable snake_case and follows a verb-object-ish convention. There is no second tool to establish or violate a pattern, so consistency is trivially satisfied but unverifiable.
One tool is borderline thin even for a narrowly scoped search server; reference discovery is offloaded to MCP Resources rather than tools. It is defensible for a single-purpose server, but a search plus retrieval pair would feel better scoped.
Search is covered, but there is no tool to list available stores or fetch a full document by identifier — agents must fall back on the MCP Resources list and rely on snippets. That is a notable gap for a documentation research surface.