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

Annotations already declare destructiveHint=true, readOnlyHint=false and openWorldHint=false, so the mutation profile is covered. The description adds genuine non-obvious context about credential handling ('les secrets fournis sont chiffrés et ne sont pas renvoyés'). However it never says what happens to an existing connection's config when re-configuring, which is precisely the destructive risk the annotation flags.

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?

Three short sentences, front-loaded with the action and followed by the security note and the cross-reference; nothing is padding. The only inefficiency is that the most actionable hint (where to find config fields) is buried last rather than attached to the config parameter discussion.

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 destructive, three-parameter mutation tool with no output schema and 0% schema coverage, the description covers what is done, that secrets are encrypted and not returned, and where to look up field names. It omits overwrite semantics, provider-specific requirements, and any indication of what the call returns or how failures surface, which leaves it thin for a destructive operation.

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% across three required parameters, so the description must compensate and largely does not. 'Configurer ou désactiver' loosely maps to the enabled boolean, but the provider enum values and the free-form config object (additionalProperties, nested) get no explanation beyond a cross-reference to describe_connections. That pointer is useful but leaves the semantics of the call itself undocumented.

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 pair and resource: 'Configurer ou désactiver une connexion du projet', which tells an agent it both upserts and toggles a connection. It is clearly distinguishable from siblings like remove_connection, list_connections and describe_connections. It stops short of naming the supported providers, which the enum in the schema has to carry.

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 phrase 'désactiver' implies using the enabled flag rather than calling remove_connection, and 'Voir describe_connections pour les champs' routes the agent to the right sibling before configuring. There is no explicit when-not guidance (e.g., when to prefer remove_connection) and no statement of prerequisites such as admin rights on the project.

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