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

Annotations already declare readOnlyHint=false, destructiveHint=false and openWorldHint=false, but the description adds genuinely new behavior: it operates only on a draft (public site untouched) and returns a preview URL. It also discloses the optimistic-concurrency requirement (expectedVersion) for corrections. It does not say whether resubmitting replaces prior proposals or what happens on version conflict, which keeps it below the top score.

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?

Four short sentences, front-loaded with the action and size cap, then the format pointer, the correction rule, and the side-effect/return statements. No filler, though the 'describe_site_projects' pointer and the correction clause could be ordered more tightly for scanning.

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 6-required-parameter mutation with no output schema and zero schema descriptions, the description covers purpose, page cap, correction protocol, non-effect on the live site, and the returned preview URL. It never addresses requestId (an 8–128 char idempotency-style key) or the role of expectedProjectVersion at the project level, leaving the two most opaque required inputs unexplained.

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 0% and all six parameters are required, so the description must carry the burden. It explains the pages limit (20, matching maxItems), the per-page correction protocol requiring targetPageId and expectedVersion for each existing page, and the draft/preview return. It leaves only the concurrency semantics of the two version fields implicit.

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 ('Déposer directement') and resource ('une proposition composée par MonAssocie') plus a concrete scope cap of 20 native editable pages. It names describe_site_projects for formats but never distinguishes itself from the very close siblings request_site_proposal and apply_site_proposal, so an agent cannot fully separate the proposal lifecycle stages from the description alone.

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?

Provides a useful prerequisite pointer ('Lire describe_site_projects pour le format') and a correction rule (targetPageId + expectedVersion per existing page), and implies draft-only status via 'Ne modifie pas le site public'. It gives no explicit when-to-use/when-not guidance relative to request_site_proposal, apply_site_proposal, or publish_site_project, so the routing decision is left to inference.

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