Enhance a TouchDesigner build (LLM-planned)

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

Beyond annotations (readOnlyHint=false, destructiveHint=false), the description discloses important safety behavior: preview-only defaults, native Apply/Keep approval, CAS/readback, compensating restore, and that autoApply never bypasses approval. This is rich behavioral context not present in the annotations alone.

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 description is concise, with the core purpose front-loaded in the first sentence. The second sentence packs many technical safeguards into a dense list, which is efficient but potentially jargon-heavy for an agent not familiar with terms like CAS/readback.

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?

With 7 parameters, a nested visualCritique object, and an output schema, the description plus schema covers the essential context. It explains the LLM-driven workflow and key safety constraints, though it doesn't clarify prerequisites like 'configured LLM' or define 'allowlisted' in concrete terms.

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 100%, so the schema already explains all 7 parameters. The description adds some high-level context (e.g., visualCritique uses a calibrated receipt), but it doesn't meaningfully extend parameter-level semantics beyond what the schema provides.

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 description clearly states the tool runs score_build and asks a configured LLM for bounded, allowlisted improvements, which is a specific verb+resource. It distinguishes itself from siblings like score_build and optimize_performance by framing itself as the LLM-planned enhancement layer.

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 schema's focusCriterion param explicitly excludes errors/perf and points to summarize_td_errors / optimize_performance, giving clear alternative guidance. The description also sets preview-only defaults and warns about apply behavior, though it doesn't fully enumerate when to prefer this over related tools like auto_repair_loop.

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