still_gate

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

With no annotations provided, the description carries the full burden and does well: it declares 'fail-closed', names the specific metrics checked (luma spread, shadow green/cyan wash), and states the failure contract (passed=false with named metric/frame failures). It also discloses that an output_dir receives a receipt and contact sheet, which is genuine behavioral context. Missing: whether inputs are mutated, permissions required, or performance/rate characteristics.

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 failure contract is front-loaded in the first sentence, the specific checks follow, and the Args block is terse and relevant. No sentence is wasted.

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 two-parameter validation gate with an output schema (which handles return shape) and no annotations, this is largely complete: it explains the failure contract, checked metrics, and the artifact written. It could still say what constitutes a pass and whether inputs must be pre-graded, but the essentials are present.

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%, so the description must compensate. It documents both parameters with format and intent: inputs as absolute still paths forming the package, output_dir as the absolute directory receiving the receipt and contact sheet. This is meaningful, but it doesn't clarify ordering, minimum count, or accepted image formats for inputs.

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: a fail-closed cohesion gate that checks luma spread and shadow green/cyan wash on a still package. It clearly distinguishes itself from siblings like still_grade or still_match by framing itself as a validation gate rather than a transformation. It falls short of a 5 only because 'cohesion gate' and 'still package' are somewhat internal jargon not fully anchored to a sibling it is not.

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?

The description implies usage (validating a package before acceptance) and mentions failure semantics, but it never states when to reach for this tool versus still_match, still_grade, or still_package. A reader can infer it is a pre-acceptance check, but no explicit when/when-not guidance or alternative is named.

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