evaluate_prompt

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

With no annotations, the description carries the full burden and does well: it discloses that deterministic gates run first and can short-circuit the job (done=true), and it names the payload fields the caller will work with (posting, profile, instructions, extracted_schema, mapped_schema). It does not state side effects, idempotency, or permission requirements, which is a meaningful gap for a tool that presumably runs and records work.

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?

The critical information — that this is step 1, that gates may short-circuit, and that the caller must finish via evaluate_submit — is front-loaded in three short lines. Minor waste in phrasing like 'everything you need to evaluate a posting with your own model', which is vague compared to the rest.

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?

There is no output schema, yet the description effectively enumerates the returned fields and describes the two sub-tasks (requirement extraction and evidence mapping) the caller must perform, which is what an agent needs to act correctly. It leaves the shape of extracted_schema and mapped_schema unexplained, forcing some 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% for the single job_id parameter, so the description must compensate; it implies job_id selects the posting/bundle returned but never states format, whether it must be an existing queued job, or what happens for an invalid id. The parameter is largely self-explanatory, so this is an acceptable but not additive result.

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 names a specific verb and resource (evaluate a posting), positions it as step 1 of an agent-mode pipeline, and distinguishes it from evaluate_submit by stating that it extracts requirements and maps evidence before submission. An agent can tell exactly what this tool does without opening the 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 clearly states the entry condition (call this first when evaluating) and the exit condition (if a deterministic gate fails, done=true and no work is required), plus the explicit follow-up (call evaluate_submit). It stops short of naming alternatives like job_evaluate or agent_run or explaining when those would be preferred over this pipeline.

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