ADR

Tool annotations

No tools have read-only/destructive annotations

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

All four hints declared on every tool

795/795 tools missing one or more hints — create_mcp_server (missing: readOnlyHint, destructiveHint, idempotentHint, openWorldHint); list_templates (missing: readOnlyHint, destructiveHint, idempotentHint, openWorldHint); get_server_info (missing: readOnlyHint, destructiveHint, idempotentHint, openWorldHint), +792 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.

Descriptions match behaviour

1 tool describes read intent but its handler mutates — conditional_prompt (line 108: exec(function_string, globals(), local_namespace))

Rename the tool, rewrite the description, or move the side-effect into a separate clearly-named tool.

Tool inputs are validated

685/795 tool handlers declare input schemas (86%)

Declare an inputSchema with zod/joi/yup on every tool definition.

Tool handlers catch errors

Only 18/795 tool handlers wrap calls in try/catch (2%)

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

No hardcoded API keys

1 live-looking API key in source: 1 Stripe secret (live)

Move secrets to environment variables (process.env.X) or your secret manager.

Tests exist

No test files found

Add tests that exercise each declared tool.

Shell command execution

6 child_process calls — runs shell commands

Prefer library functions over shell-outs. If you must shell out, ensure all inputs are properly escaped.

Tool description accuracy

conditional_prompt: description implies read-only but handler writes/deletes/executes

Update tool descriptions to accurately reflect all capabilities — especially write, delete, or execute operations.