Unified Run Comparison

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

readOnlyHint=false is consistent with the description's disclosure that the comparison is stored, so no contradiction; the persistence side-effect is stated beyond the annotation. However, it does not say where state is written, whether it overwrites a prior comparison, or what is returned, which matters for a mutating tool with no 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?

Two tight sentences, action and outcome front-loaded, with the release_check linkage in the strongest position. The trailing 'Same behavior as lens.compare_runs' is genuinely useful routing information rather than padding.

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 four-parameter mutation tool with no output schema and 25% schema coverage, the description omits the input payload shape, threshold semantics, where the stored comparison lives, and what the caller gets back. It names the release-gating role but leaves the mechanics an agent needs to invoke it correctly undefined.

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?

Only session_id carries a schema description (25% coverage), and the description adds nothing about baseline, candidate, or regression_threshold. The two most important inputs are untyped arrays of arbitrary objects with no documented element shape, so an agent must guess what to pass and how the threshold is interpreted.

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?

Names a specific verb and resource pair (compare baseline and candidate trace runs) and states the outcome (stores the comparison for core.release_check), which an agent can act on. The sibling differentiation is weak: 'Same behavior as lens.compare_runs' asserts equivalence rather than telling the agent which namespace to pick, so it informs but does not distinguish.

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 by 'for regressions' and the downstream link to core.release_check, so an agent can infer this belongs in a release-gating flow. There is no explicit when-to-use/when-not guidance and no statement about why an agent would choose core.compare_runs over lens.compare_runs when they behave identically.

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