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

Annotations already provide readOnlyHint: true and destructiveHint: false, making the read-only nature clear. The description adds valuable behavioral context beyond annotations: it confirms this tool never accesses the machine and performs no action, and states the return is the complete response in read-only mode. It also warns against adding any follow-up actions, which is not in annotations.

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 two sentences: the first defines the trigger conditions precisely, the second sets strict post-call behavior rules. No wasted words; every sentence is necessary and front-loaded with critical usage conditions.

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?

Given the tool has 2 optional parameters, a full output schema (not provided but exists per context), and comprehensive annotations, the description is fully self-contained. It explains purpose, triggers, safety profile (read-only, no machine access), and post-call constraints. No gaps remain for the agent to infer incorrectly.

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%, so the description must compensate. While the description does not explicitly list parameter names or formats, it implies 'target' refers to what to install/scan/delete (e.g., Ollama, model, driver) and 'requested_action' maps to the listed actions (install, download, etc.). The description's trigger list directly corresponds to the action enum, providing meaningful context beyond the schema. The 'target' maxLength and default are schema-only, but the description clarifies its semantic scope.

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 clearly specifies the verb 'expliquer' (explain) and the resource 'la limite des actions locales' (local action boundary). It lists exact trigger conditions (install, download, update, run, delete, scan, benchmark) that distinguish it from sibling tools like 'check_pc_for_local_ai' and 'recommend_runtime'.

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 description explicitly states when to call the tool: for any request to install, download, update, run, or delete Ollama, a model, or driver, as well as scanning or benchmarking. It also provides strong negative guidance: after the call, do not add commands, procedures, web searches, external sources, or manual alternatives. This covers both triggers and post-call constraints.

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