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

Annotations already declare readOnlyHint=false, destructiveHint=false and idempotentHint=true, so the safety profile is partially covered. The description adds non-obvious consequences the annotations cannot express: cancel leaves already-created drafts in place, and retry is scoped only to failed pages rather than restarting the import. That side-effect disclosure is the main value-add, consistent with destructiveHint=false.

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?

A single semicolon-delimited sentence that front-loads the action name before its effect, with zero filler. The parenthetical about drafts is the only elaboration and it is the one detail an agent could not otherwise infer.

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 two-parameter control tool with no output schema, the definition covers the state machine that governs correct invocation (completed vs in-progress import) and the side effect of cancel. What remains unstated is what retry returns (a job identifier, a count of retried pages) and whether the tool errors when the import is in the wrong state.

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 carry the meaning, and it fully disambiguates the two enum values of 'action' with distinct state preconditions. The 'id' parameter receives no explanation at all (it is only bounded to 1-64 chars in the schema), leaving the agent to infer that it identifies the import to control.

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 pairs each action with a concrete outcome: retry re-runs only the failed pages of a finished import, cancel stops an in-progress import. The resource itself ('website clone import') is carried by the tool name and title rather than the description, so the agent must combine both to know precisely what is being controlled, which keeps it just short of a 5.

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?

Selection conditions are stated explicitly for both enum values and are mutually exclusive by state: retry applies to a 'import terminé' (completed import), cancel to a 'import en cours' (in-progress import). That is genuine when-to-use guidance, though no alternative sibling is named (e.g. get_website_clone or cancel_site_project_job) for cases where the agent needs a different operation.

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