Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

Annotations already flag destructiveHint=true and openWorldHint=true, but the description adds the substantive detail those hints cannot: recreate deletes the existing connector, an interactive ChatGPT window may open for sign-in, credentials are never handled or passed as an argument, Developer mode is enabled when possible, and the run returns run_id plus per-stage statuses. This is exactly the context that keeps an agent from misinterpreting a paused run as a failure.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Is the description appropriately sized, front-loaded, and free of redundancy?

Front-loaded with the core action, then constraints, then the destructive flag, then the return shape — a sensible ordering. Nearly every sentence carries distinct operational value; it is dense but not padded, though the sign-in explanation runs slightly long for a tool with only three optional parameters.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

For a mutation tool with destructive and open-world hints, it covers the prerequisites (Developer mode), the auth handoff (user signs in themselves), the recovery path (recreate for token rotation), and the polling contract (run_id, wait_seconds, needs_user). An output schema exists, yet the brief return summary is still helpful and nothing an agent needs to call this correctly is missing.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema coverage is 100%, so baseline is 3; the description still adds meaning by explaining recreate as a destructive delete-and-rebuild for token rotation and by framing needs_user stages in the run output. That said, wait_seconds behavior is left entirely to the schema, so it is not a full superset of parameter guidance.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Does the description clearly state what the tool does and how it differs from similar tools?

States a specific verb and resource ("Connects this Mac to ChatGPT: creates (or repairs) the LMCP connector inside the user's ChatGPT account") and explains the resulting capability (ChatGPT can use this Mac's tools). That separates it cleanly from siblings like connect_m365_account, connect_servicenow, or lmcp_install_upgrade, which are per-service connectors rather than the ChatGPT-side connector.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Does the description explain when to use this tool, when not to, or what alternatives exist?

Gives clear conditions: run "on this Mac only", pass recreate:true for token rotation, and treat a stage with status needs_user as waiting-for-the-person rather than broken. It stops short of naming an explicit alternative (e.g. disconnect/reconnect paths or lmcp_doctor when setup is already broken), so it is strong context without full when/when-not routing.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.