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

Annotations already cover safety posture (readOnlyHint true, destructiveHint false, openWorldHint false), and the description adds substantial behavior beyond them: deterministic rules vs. language model, millimetre precision, cascading edits (attached walls extended, furniture followed, obstacles pushed, switches relocated), and a validation gate that rejects defective walls and forbids moving a load-bearing wall 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?

Front-loaded with the WHEN trigger and written as a dense but well-ordered chain (trigger → transformation → safeguards → return shape → downstream). It is long and semicolon-heavy without bullet structure, but nearly every clause carries information an agent needs.

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 no output schema, the description compensates by enumerating the full return object ({compris, source, operations, journal, erreurs, nouvelles_alertes, valide, scene, difference}) and by explaining validation, deterministic fallback, and downstream routing. Nothing required to invoke or interpret the tool 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 100%, so the baseline is 3, but the description adds meaning the schema lacks: it points to the operation catalogue endpoint (GET /api/v1/cao/copilote) and specifies that supplied operations are 'applied without the language model', clarifying the relationship between the `operations`, `instruction` and `modele` parameters.

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 specific verb and resource ('change an existing building scene') and enumerates concrete instruction types (move a wall, widen a door, rotate furniture) that map to the typed operations the tool produces. It also positions itself against siblings by noting the resulting `scene` is forwarded to cao_generer_ifc or cao_pdf, so an agent can distinguish editing from generation.

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 'WHEN a user asks to change an existing building scene in plain words' clause gives an explicit trigger with representative French utterances. It implies generation is a separate sibling but never states when NOT to use it, nor names an alternative editing/validation tool such as cao_verifier.

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