TakoData/tako-mcp

Tool annotations

8/17 tools have annotations

Add readOnlyHint or destructiveHint annotations to every tool so hosts can warn users before invoking.

All four hints declared on every tool

17/17 tools missing one or more hints — search_tako (missing: readOnlyHint, destructiveHint, idempotentHint, openWorldHint); web_search_tako (missing: readOnlyHint, destructiveHint, idempotentHint, openWorldHint); deep_search_tako (missing: readOnlyHint, destructiveHint, idempotentHint, openWorldHint), +14 more. OpenAI's directory rejects tools where any of the four hints are missing or non-boolean.

For every tool, set all four hints (readOnlyHint, destructiveHint, idempotentHint, openWorldHint) to explicit true/false values that match the handler’s actual behaviour.

Destructive tools are labelled

1 tool perform destructive updates without destructiveHint — upload_file_to_visualize deletes at line 186 (os.remove(temp_file_path))

Add destructiveHint:true to any tool whose handler calls .delete(), .upsert(), .update(), unlink, rm, DELETE, DROP, REPLACE INTO, or any operation that overwrites existing data.

openWorldHint matches behaviour

6 tools claim openWorldHint=false but make outbound HTTP — knowledge_search → dynamic (line 114); get_chart_image → dynamic (line 204); get_card_insights → dynamic (line 266)

Either flip openWorldHint to true (acknowledging external reach) or refactor the handler to avoid outbound HTTP. If the only network calls are to a known fixed endpoint that you consider part of the tool itself, document that and keep openWorldHint=false; otherwise true is honest.

Tool handlers catch errors

13/17 tool handlers wrap calls in try/catch (76%)

Wrap each tool handler body in try/catch and return a structured error response.

Tool test coverage

Only 4/17 tools referenced in tests (24%)

Write tests that reference each tool by name so every tool has at least one test.