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

The annotations declare readOnlyHint=false and openWorldHint=true but say nothing about cost, so the 'uses points, metered on the models used like the website' note is genuine added context an agent needs before firing a paid call. The async implication (answers must be fetched separately with get_deep_thread) is also disclosed, which annotations do not cover.

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?

Two tight clauses front-load the action and the read path, and the cost caveat is compact. The trailing 'like the website' is slightly loose phrasing but not wasted length.

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 four-parameter, no-output-schema mutation, the description covers what it does, what must exist first, where results are read, and what it costs. The only meaningful gap is whether the query is synchronous or queued and how many committee seats respond by default.

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 75% and the schema already documents analysis_id and idempotency_key. The description corroborates 'follow-up question' for message and 'finished run' for analysis_id but says nothing about the optional mentions/committee-seat parameter, leaving the model-mixing behavior described only in the schema.

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?

The description states a specific verb and resource — 'ask a follow-up question about a finished deep research run' — and routes the retrieval half to the sibling get_deep_thread. That makes it distinguishable from get_deep_analysis (fetch a run) and start_deep_analysis (begin a run) without opening any schema.

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?

It gives a clear precondition ('a finished deep research run') and names get_deep_thread as the way to read the answers, so the agent knows the ask is not where results come back. It doesn't explicitly say what to do instead for an unfinished run (e.g., deep_resume or start_deep_analysis), which keeps it short of a 5.

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