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

Beyond the annotations it discloses real behavioral traits: the source file and model are untouched, the exported header carries a beta mark that must be surfaced to the user, and compare_to validation can come back "schema checked false" (too_large/time_budget) which means unchecked, "never clean." That last point is precisely the caveat an agent needs to avoid over-reporting a clean export, and it is consistent with destructiveHint=false.

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 purpose and non-destructiveness guarantee are correctly front-loaded, but the middle sentences are compressed into near-telegraphic fragments ("export_ifc's own compare_to adds the balance with schema findings (none without it)", "a whole goes with its parts"). The information is valuable yet the phrasing forces re-reading, so density partly defeats conciseness.

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 write-a-file tool with no output schema, it covers the mutation boundary (model unchanged), the return value (file location), the trustworthiness of validation, subsetting semantics, and environment limits. Nothing an agent needs to call it correctly and report results honestly is missing.

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 already 100%, so the baseline is 3; the description earns above that by explaining what compare_to actually changes (adds schema findings to the balance) and what guids/set scoping preserves (structure, types, values outside the subset, and parts of a whole via subset.added). It does not add anything about `except` or the default `revision`, which remain schema-only.

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?

"Writes a revision as an .ifc file and returns its location" names a specific verb, resource, output artifact, and return value, and immediately adds "the original file is never touched." No sibling exports IFC, so the tool is unambiguous against the rest of the catalog.

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 prescribes a concrete pre-handover workflow ("check_model, then get_history compare_to the revision you started from") and states a scope constraint ("Local files only"). It stops short of saying when to skip this tool or what to do if the location is not writable, but the positive guidance is specific.

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