MCP Tools
ask
Search across all your knowledge with a single natural language question
Search across all your knowledge — documents, memories, conversations, and insights — with a single natural language question. Returns structured results with citations. This is the recommended entry point for querying Textrawl.
Parameters
| Parameter | Type | Required | Default | Description |
|---|---|---|---|---|
question | string | REQUIRED | - | Natural language question to search your knowledge base (1-10000 chars) |
scope | enum | OPTIONAL | auto | Which sources to search: auto, documents, memory, conversations, insights |
limit | number | OPTIONAL | 10 | Maximum results per source (1-50) |
Example Request
Auto Scope (Default)
{
"question": "What did we decide about the migration strategy?"
}Scoped to Documents
{
"question": "API rate limiting best practices",
"scope": "documents",
"limit": 5
}Scoped to Memory
{
"question": "What do I know about Alice's role?",
"scope": "memory",
"limit": 20
}Response
{
"question": "What did we decide about the migration strategy?",
"totalResults": 8,
"counts": {
"documents": 3,
"memories": 2,
"conversations": 2,
"insights": 1
},
"results": [
{
"type": "document",
"score": 0.89,
"documentId": "abc-123",
"documentTitle": "Migration Plan Q1",
"content": "We decided to use a phased migration...",
"sourceType": "note"
},
{
"type": "memory",
"score": 0.82,
"entityId": "ent-456",
"entityName": "Database Migration",
"entityType": "project",
"content": "Phased approach approved in January review",
"source": "conversation"
},
{
"type": "conversation",
"score": 0.78,
"sessionId": "sess-789",
"sessionKey": "migration-planning",
"title": "Migration Planning Session",
"summary": "Discussed phased migration strategy..."
},
{
"type": "insight",
"score": 0.71,
"insightId": "ins-012",
"insightType": "connection",
"entities": ["Database Migration", "Infrastructure Team"]
}
]
}Results are truncated: document content to 500 characters, conversation summaries to 300 characters.
Error Responses
| Error | Cause | Fix |
|---|---|---|
Database not configured | Missing database connection | Set DATABASE_URL |
Embedding provider not configured | Missing embedding API key | Set OPENAI_API_KEY, configure Ollama, or set GOOGLE_AI_API_KEY |
Failed to search knowledge base | Database or embedding error | Check connectivity and logs |
Notes
- This tool is read-only and idempotent — it does not modify any data.
- When
scopeis set toauto, Textrawl searches all enabled sources automatically (documents, plus memory, conversations, and insights if their respective feature flags are enabled). It does not rank or limit to a smaller subset. - Results include citations back to the original source, making it easy to verify or dive deeper.
- For more control over search weights and filtering, use the
searchtool directly.
Related Tools
search- Advanced hybrid search with weight tuning and filtersquery_memory- Search entity memories directlyquery_conversations- Search past conversations directlyget_insights- View discovered insights