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

Annotations already flag destructive and idempotent, but the description goes well beyond them: it enumerates what is destroyed (runs, pages, change records, webhook history), notes an in-progress run stops, that already-spent credits are not refunded, that a second call returns 404, and that an admin API key is required with OAuth apps refused via admin_only. This is exactly the extra context annotations cannot carry.

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 purpose and constraint, then grouped into Inputs and Effect blocks that are easy to scan. Slightly long, but nearly every sentence carries a distinct fact (alternatives, guardrail, destruction set, auth requirement).

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, guarded mutation with rich annotations and an output schema, the description covers usage, inputs, side effects, auth, and error behavior. Nothing an agent needs to call this safely is missing.

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 coverage is 100%, so the baseline is 3, but the description adds the consequence-level meaning that confirm_name must match exactly 'or nothing is deleted' and that project_id comes from list_projects. That guardrail framing is real value over the schema's field-level text.

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 and resource ('Delete a project for good, with everything its runs stored') and immediately scopes the blast radius. It is unmistakably distinct from siblings like cancel_job and update_project, which it names.

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?

Gives an explicit when-to-use trigger ('only when the person asks for the project to go, e.g. to free a slot after create_project is refused with plan_limit') plus two named alternatives with the conditions that select them (cancel_job to stop a run, update_project with schedule='manual' to stop scheduled runs). Nothing is left to inference.

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