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

With no annotations, the description carries the full burden and does most of it: it discloses that the no-job mode is side-effect free (no compile submitted), enumerates terminal statuses (running/succeeded/failed/cancelled), explains cancellation provenance, and warns against treating a lagging last_success_seq as proof of compilation. The remaining gap is the non-terminal payload shape beyond the four states, which is partly covered by the output schema.

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 mode split and structured around it, so the agent hits the decision point first. It is dense and runs long, with a couple of clauses (engine, state_seq bookkeeping detail) that could be tightened, but essentially every sentence carries operational instruction rather than filler.

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 single-parameter tool with an output schema, the description supplies everything an agent needs to invoke it correctly and interpret results: which mode to pick, which statuses are terminal, what cancelled means, and the is_fresh_green caveat that prevents a wrong conclusion. Nothing material is left to inference.

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 0% — job_id has no description at all — so the description must compensate, and it does: it defines the parameter's two states (absent vs present), its origin ('the asynchronous compile job returned by the compile tool'), and what each state changes in behavior and return value. That is more meaning than the schema could ever convey.

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?

Opens with 'Two modes' and then specifies exactly what each returns: WITHOUT job_id the latest compile state (explicitly noting no compile is submitted), WITH job_id the polled status of an async job. It names the sibling tools compile and compile_log, so an agent can place it in the workflow 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?

Gives explicit ordering ('use this FIRST when you need to know whether the project compiles'), an explicit pairing ('pair it with compile_log for the raw error log'), and an explicit stop condition for the cancelled case ('do not keep polling or retry the compile'). It also bounds polling at roughly two minutes, which is a when-to-stop rule rather than generic advice.

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