bonito-mcp-server

# bonito-mcp The **killer differentiator** of Bonito (BON-004): an MCP server that gives any AI agent (Claude, Bedrock, Strands) native access to deep database observability. No Datadog DB Monitoring, DBmarlin or Percona PMM expose MCP — Bonito does. It is a thin wrapper over the `bonito-api` REST layer (business logic lives in store/API, not duplicated here). ## Run ```bash uv pip install -e bonito-mcp # local (repo checkout) pip install bonito-mcp-server # from PyPI export BONITO_API_URL=http://localhost:8100 export BONITO_API_KEY=change-me bonito-mcp # stdio transport ``` > **PyPI name note:** the distribution is `bonito-mcp-server` — the bare name > `bonito-mcp` is already taken on PyPI by an unrelated project. The console > command is still `bonito-mcp`. With `uvx`, use > `uvx --from bonito-mcp-server bonito-mcp`. ## Connect to an agent **Claude Desktop** — `claude_desktop_config.json`: ```json { "mcpServers": { "bonito": { "command": "bonito-mcp", "env": { "BONITO_API_URL": "http://localhost:8100", "BONITO_API_KEY": "change-me" } } } } ``` **Claude CLI**: ```bash claude mcp add bonito -- bonito-mcp ``` **Bedrock / Strands:** the same server runs over stdio/SSE — wire it as a tool provider and the agent gets all 8 tools. ## Tools | Tool | What the agent asks | |------|---------------------| | `get_slow_queries` | "which queries are slow?" | | `get_active_locks` | "are there locks right now?" | | `get_blocking_tree` | "who is blocking whom?" | | `explain_query` | "what's the execution plan for this query?" | | `get_wait_analysis` | "what are the sessions waiting on?" | | `compare_to_baseline` | "is this query slower than usual?" | | `suggest_remediation` | "what should I do about a lock / slow query / bloat?" | | `get_db_health_summary` | "give me a health summary of the database" | ## Config | Env var | Default | Description | |---------|---------|-------------| | `BONITO_API_URL` | `…