inspect_deposit

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

The description does not contradict the readOnlyHint annotation; actually, 'Relire' (re-read) aligns with a read-only operation. It goes beyond the annotation by specifying what is read: 'brouillon exact, son empreinte, son test et le journal d'envoi' (the exact draft, its fingerprint, its test, and the sending log). This provides useful context about the type of data involved, but it is not exhaustive. With the annotation covering read-only behavior, the description adds detail on the outputs, which is a plus.

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 description is a single sentence in French, which is concise and front-loaded with the purpose. It is efficient with no wasted words. However, it is in French, which might be appropriate if the system expects French, but it could be a barrier for English-based agents. Still, length is appropriate and structure is clear.

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?

The description is a single sentence in French, which is concise and front-loaded with the purpose. It is efficient with no wasted words. However, it is in French, which might be appropriate if the system expects French, but it could be a barrier for English-based agents. Still, length is appropriate and structure is clear.

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?

The schema has only one parameter, draft_id, with no description, and the schema description coverage is 0%. The description does not mention draft_id at all, so it does not add any meaning to that parameter. However, with only one parameter and a clear name 'draft_id', the agent can infer it's an identifier for the draft. Since the parameter is obvious, a baseline of 3 is appropriate because the schema itself is minimal and the description doesn't hinder understanding.

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 is a French sentence: 'Relire le brouillon exact, son empreinte, son test et le journal d'envoi.' Translated, it means 'Re-read the exact draft, its fingerprint, its test, and the sending journal.' It uses a specific verb ('Relire' - re-read) and resource ('brouillon' - draft). It indicates the tool inspects a draft, which is distinct from validate, submit, etc. However, it is vague about what 'inspect' entails and does not clearly name the action in English or connect to the tool's name. It could be interpreted as just viewing the draft, but the mention of fingerprint, test, and sending log clarifies it's an inspection of those artifacts. Still, the purpose is not fully clear because it doesn't state what the output is or the exact scope.

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?

There is no explicit guidance on when to use this tool versus the siblings like validate_deposit or prepare_deposit. The description implies it's for checking a draft before submission, but does not state when to inspect versus when to validate. Given the context of sibling tools (validate, submit, prepare), it would be helpful to specify that this is for reviewing the exact state before actions like validate or submit, but this is missing.

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