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

Annotations already declare readOnly, idempotent, non-destructive and closed-world, so safety needs no restating; the description instead spends its space on domain behavior the agent could not otherwise know: the corpus contains no named domains, numbers or persons, and a cited organization (e.g. ANTAI, a bank) is typically the impersonated party, not the scammer. That impersonation caveat directly prevents a serious misinterpretation of results.

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 what the tool is, followed by the negative scoping, the impersonation caveat and the redirect — a logical order with no filler sentences. The prose is dense and clause-heavy in French, which costs a little readability, but every sentence carries distinct information.

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 7-parameter, zero-required search tool with no output schema, the description adequately establishes the corpus and its limits, which is the main correctness risk here. What remains uncovered is operational detail — pagination via limit/cursor and the effect of combining filters — which the schema handles only partially.

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?

With 57% schema coverage and 7 parameters, the schema already documents the three enums and include_guides. The description only indirectly touches parameters — 'TYPES d'arnaques' hints at the family dimension and 'guides « Je suis victime »' maps to include_guides — but adds no meaning for query, limit or cursor, so it does not compensate for the uncovered half.

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 names a specific verb and resource (search within IP73's scam guide: explanatory sheets on scam TYPES in France plus 'Je suis victime' guides) and, critically, defines the corpus by negation — 'Ce ne sont PAS des signalements', no lists of fraudulent domains/numbers/people. That negative scoping lets an agent distinguish it from a generic content search or a fraud-lookup tool without opening the schema.

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 states a clear when-not-to-use condition (verifying a site or filing a report) and routes those needs to specific official channels (17Cyber, Pharos, Signal Conso), which is unusually actionable guidance. It does not, however, compare itself to sibling tools such as search_content for general lookups, so the boundary against in-catalog alternatives is left implicit.

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