detect_faces

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

Annotations only declare it is a non-read-only, non-destructive, closed-world call; the description goes well beyond that by disclosing the asynchronous job model, the side effect of writing contact.jpg, the grid labelling convention ('face N' / 'id N: <appearances>x', '*' for unverified references), and that reference crops are verified to re-detect as one face. That is exactly the extra context annotations cannot carry.

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?

Front-loaded with the action and next steps, then mode selection, then the shared artifact output — a logical order. It is dense with parentheticals and backticks but each clause carries information; the trailing 'view it first' instruction is slightly tacked on but still earns its place.

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, yet the description still usefully explains the `faces` vs `identities` return shapes and the crop-reference contract for FaceMapping.target_face. It falls short only on the unexplained `job_id` and `controls` parameters, which matter for a six-parameter job-oriented tool.

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 — and it does for `frame` (zero-based), `sample_frames` (N evenly spaced) and `frames` (explicit, at most 64). However `job_id` and `controls` are never mentioned anywhere, leaving a third of the parameters with no semantics at all.

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?

Opens with a specific verb+resource ('Queue face detection') and immediately distinguishes two operating modes: single-frame detection returning `faces`, versus multi-frame detection (`sample_frames`/`frames`) returning grouped `identities`. An agent can tell exactly what this produces 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?

Prescribes an explicit workflow — queue, then poll `job_status`, then inspect crops via `view_artifact` — naming the sibling tools in order. It also states the selecting condition between modes: use `frame` for a single frame, use `sample_frames`/`frames` 'to find who appears across a video', with the 64-frame cap.

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