Skip to main content

Documentation Index

Fetch the complete documentation index at: https://docs.datris.ai/llms.txt

Use this file to discover all available pages before exploring further.

Location: examples/test-mcp-server/ An integration test script that exercises the MCP server tools end-to-end. It imports the _dispatch function from server.py and calls each tool directly, testing the full round-trip through the REST API.

​
Test Flow

  1. Connectivity — verify server is reachable (get_version)
  2. Service health — check all backend services (check_service_health)
  3. Metadata discovery — list PostgreSQL databases/schemas/tables, MongoDB databases/collections, and all vector store collections
  4. CSV → PostgreSQL — create pipeline, upload CSV, wait for ingestion, inspect columns, query with SQL and natural language
  5. JSON → MongoDB — create pipeline, upload JSON, wait, query with find()
  6. Data profiling — profile a CSV file with AI
  7. Config upload — upload a validation schema
  8. Secrets — update an AI provider secret
  9. Kill job — test kill with a dummy token (graceful handling)
  10. Cleanup — delete structured pipelines
  11. PDF → pgvector — create pipeline, search, AI answer (RAG), cleanup (requires PDF test file)
  12. Summary — report pass/fail counts

​
Setup

cd examples/test-mcp-server
pip install requests python-dotenv

​
Run

python app.py
Requires Datris running via docker compose up. Natural language queries and vector search require an AI provider to be configured.