three_live_convert

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

Annotations already declare this is a non-read-only, closed-world, idempotent operation, so the safety profile is covered. The description adds useful spec detail (USDZ 64-byte alignment, per-platform target runtimes) but omits material behavior: whether compiling an inline script executes arbitrary JS, whether the binary is returned or only written when outputPath is given, and what happens on compile failure.

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 that puts the inputs and outputs first, then the platform targets. The trailing platform enumeration is dense but informative rather than filler; minor cost is that the sentence is long enough to read as a run-on.

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 4-parameter, all-optional tool with no output schema, the description covers inputs, output formats, and target runtimes well. The main residual gap is return behavior — whether the .glb/.usdz bytes come back in the response or only land at outputPath — which matters because there is no output schema to answer it.

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 coverage is 100%, including the enum for format and the code-vs-inputPath alternative, so the schema already carries parameter meaning. The description contributes nothing parameter-specific (no note on how format is inferred from outputPath, no precedence rules when both code and inputPath are supplied), leaving this at the baseline.

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 concrete verb (compile) and the exact resources involved: Three.js script, HTML document, or .glb input, producing .glb or .usdz output. The target runtimes (Filament/SceneCore, RealityKit/Quick Look, Web) make it unmistakable versus siblings three_live_validate, three_live_render, and three_live_sim, which imply checking, previewing, and simulating rather than converting.

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 only implied by the conversion framing; there is no explicit 'use this when you need a shipping asset, use validate to check first' routing. The description never names an alternative or a precondition (e.g., validate the scene before compiling), so the agent must infer when to pick this over three_live_validate or three_live_render.

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