modeling_progress

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

With no annotations, the description carries the full burden and does well: reset discards only the checklist (not the model), state is stored in the Blender scene and saved with the .blend, delivery checks are preserved across recording, and explicit limits are stated ('No image-content analysis, evidence-truth verification or complete dependency tracking'). The clipped phrasing leaves a few invariants ambiguous, keeping it below 5.

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 in the first sentence, but the body is a dense run of fragments and multi-clause sentences ('Passing stages requires their comparisons; failed comparisons require recorded repairs and fresh model images before passing') that demand re-reading and intermix rules with limitations.

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 complex stateful tool with 18 optional params and no output schema, the description covers workflow order, invariants, invalidation rules, and explicit non-goals, which is nearly what an agent needs. It stops short of describing what each action returns (beyond 'report lists blockers'), leaving some call-outcome ambiguity.

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 61%, so the schema already documents most of the 18 parameters (features, findings, comparisons, on_error, evidence, etc.). The description only adds meaning for a few behaviors (reset scope, report/ready_for_delivery semantics) and gives no per-parameter syntax beyond what the enum and schema titles convey, so the baseline 3 applies.

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 line names a specific artifact and scope: a 'Scene-persistent per-part completion checklist' with stage inspection and evidence recording. It is clearly a bookkeeping/state-machine tool, distinguishable in kind from siblings like review_model or analyze_quality, but it never explicitly names those alternatives to route the agent.

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?

Offers substantial when/how guidance: passing requires all earlier stages and no unresolved findings, rechecks invalidate later stages, changes to geometry/material/reference require invalidating the earliest affected stage first, and only one active plan per scene. It lacks explicit exclusions ('do not use this instead of X'), but the procedural context is strong.

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