diagnose_job

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

Annotations only declare readOnlyHint=true, so the description carries most of the burden and does so well: it discloses that it aggregates SLURM state and log tails, performs pattern-based cause recognition, and additionally searches for resume points on timeout. It omits cost/latency and how deep it scans, but the behavioral picture is substantially richer than the annotation.

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?

Four compact sentences, front-loaded with the single-call promise and the list of recognized causes. Every clause adds signal; only the enumeration is slightly long, which is justified by the router value it provides.

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?

An output schema exists, so return structure need not be explained, and the failure-mode coverage is thorough for a read-only diagnostic. The only material hole is parameter guidance, which the 0%-coverage schema pushes back onto the description and which is not addressed.

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 0% for two parameters, so the description must compensate — and it does not. The 'lines' parameter (log tail length, default 80) and the format/expectation of 'job_id' are never mentioned, leaving the agent to infer their meaning entirely from property names.

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+resource (diagnose/autopsy a failed job) and enumerates the exact aggregates it produces: SLURM state, log tails, recognized causes and remedies. The named failure modes (x86 binary on aarch64 node, VRAM saturation, OOM, bus error, NCCL failures, quotas) clearly separate it from siblings like job_output or job_status.

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?

Explicit trigger and priority: 'call as soon as a job fails, before any other investigation,' plus a conditional extension for time-limit cases. It does not, however, name the alternative tools (job_output, job_status) an agent might otherwise reach for, so the routing is implied rather than spelled out.

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