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

Annotations already declare readOnlyHint/idempotentHint/non-destructive, and the description adds real context beyond them: the log includes agency-approved proposals, results are only measured 'once there has been time', and a before/after leads comparison is included. It does not cover pagination or result volume limits, which would have pushed it to a 5.

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 resource and its contents, then the routing clause, then the read-only note. The four quoted example intents are slightly redundant but each maps to a distinct question the log answers, so the length is defensible.

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?

With no output schema, the description carries the return-value burden and does so well, enumerating the log's fields (timing, change, rationale/evidence, expected vs. measured result). Combined with the routing clause and read-only statement, an agent has everything needed to call it correctly except the 'days' semantics.

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 description coverage is only 50%: 'client' is documented in the schema (and echoed by the description's 'for a client'), but 'days' (default 90, range 7-365) is undocumented in both schema and description. The description's 'when' hints at a time dimension but never explains the lookback window, so it fails to compensate for the coverage gap.

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 resource (the Autopilot change log) with the exact fields it exposes: when, what, why/evidence, expected result, and measured result. It explicitly distinguishes itself from the sibling autopilot_status ('For what it has learned, use autopilot_status'), so an agent can route without opening either 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?

Names concrete user intents it answers ('what has Autopilot done for this client', 'why did it do that', 'did it work') and gives an explicit exclusion with an alternative tool for the learnings use case. Nothing about when to pick this over its siblings is left to inference.

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