rig_face

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

With no annotations, the description carries the full burden and delivers unusually rich behavioral detail: it enumerates what gets constructed (turn, eyes, brows, mouth/jaw, cheeks, hair, clips, events) and the mechanics behind each. However, for a mutation tool it never says whether an existing rig is overwritten or duplicated, nor what the call returns or whether it is safe to re-run, which are meaningful gaps.

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?

Purpose is front-loaded correctly, and the layer-name list, clip list, and event names earn their place. But the build/clip paragraphs descend into implementation minutiae (Poisson micro-saccades, scaleX = scaleY^-1/2, 80 ms down / 160 ms up) that add bulk without helping an agent decide or call the tool, making it denser than needed.

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 12-parameter, no-output-schema, no-annotation tool of high complexity, the description is largely sufficient: it covers layer recognition, constructed parts, clips, events, and the most consequential parameters. It falls short on undocumented parameters and on the absence of any statement about return values or overwrite behavior.

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% across 12 parameters, so the description must compensate. It explains project (convention query), clips (default all), parent (default juice_core or root), spine (joins look_at two frames after the head), talk_text (lipsync source), and turn_range (turn range scale). It leaves seed, pitch, intensity, features, hair_physics, and idle_duration entirely undocumented, so compensation is only partial.

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 opening sentence states a specific verb (rig), resource (a whole face), and scope (from named layers in one call, then build its clips). It is clearly distinguishable from siblings like rig_mesh, rig_ik, rig_turn, and rig_biped, which each target a narrower or different resource.

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 makes the trigger condition concrete: you need layers following the documented naming convention, and face/head is REQUIRED while missing parts are skipped and reported. It also documents a query mode (project="" returns FACE_LAYERS without changing anything) and notes it internally builds rig_turn as the base layer, implying it supersedes that sibling. It never states explicit when-not-to-use or prerequisites such as an open project.

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