evaluate_agent_response

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

With no annotations supplied, the description carries the full behavioral burden. It says what is checked but not what happens on a violation (throw, return a verdict object, list of failures), whether checks short-circuit, or whether 'verified source-slice citations' means the tool fetches/validates external sources — a potentially expensive or failing operation left undisclosed.

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?

A single front-loaded sentence with a colon-delimited enumeration — efficient and scannable, with no filler. Density borders on terseness; the facet labels are compressed enough that a reader must infer what 'verified' and 'anchor line' mean operationally.

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?

The tool has two top-level params but a four-field nested contract, no annotations, and no output schema, so the description must cover inputs and outputs alone. Inputs are reasonably covered; the return value — the entire point of an evaluation/check tool — is completely unspecified.

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 50%: only 'response' is documented in the schema, while none of the contract's nested properties (anchor, schema, source, forbiddenClaims) are. The description partially compensates by mapping the four facets to those fields and hinting that the anchor is a line and that source slices are verified, but it gives no format details for anchor, schema, or forbiddenClaims entries.

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 and resource ('Check one agent response against a contract') and enumerates the four contract facets it evaluates: anchor line, JSON payload schema, forbidden claims, and source-slice citations. The scope word 'one' signals single-response evaluation, distinguishing it from batch-oriented siblings like dedup_and_lint_dossier, though no sibling is named explicitly.

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?

Usage is implied rather than stated: 'Check one agent response against a contract' suggests a per-response validation call, likely after an agent has produced output. There is no explicit statement of when to prefer this over dedup_and_lint_dossier, no preconditions, and no note on whether the response must already exist or be retrievable.

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