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

Far exceeds the annotations: discloses the required permission (items.delete), irreversibility, organization scoping, and that unknown ids and cross-org ids return the same error (a deliberate information-leak guard). The two-step confirm_token mechanics — first call writes nothing, second call with the same args plus token performs the write, token is single-use with a 5-minute TTL, a `confirm` boolean is ignored — is exactly the behavioral context annotations cannot convey.

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?

Dense but front-loaded: permission, action, irreversibility, then the confirm flow. Nearly every sentence carries safety-critical information, though 'does not write, delete, cancel, or send' is slightly over-enumerated for what is essentially 'this call is a no-op.'

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?

With no output schema, the description covers the first call's response shape (confirmation_required, summary, confirm_token), which is the part an agent most needs. It does not describe what the second, committing call returns, leaving a small gap for a destructive tool.

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 load, and it does: pdf_id is constrained to one value per call, no wildcards, must belong to the organization; confirm_token is fully specified in origin, lifetime, single-use semantics, and the requirement to repeat identical arguments. It even preempts a likely agent mistake by noting a `confirm` boolean is ignored.

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 ('Permanently delete one stored PDF by pdf_id') with explicit singular scope, and distinguishes itself from bulk siblings by declaring 'Wildcards and bulk deletes are not accepted.' An agent can tell this apart from bulk_delete_items without opening either schema.

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 a clear gating condition ('Call this only when the user has asked to delete that PDF') and rules out bulk/wildcard use, which implicitly routes to bulk_delete_items. It stops short of naming that sibling or describing when the two-step token flow should be re-initiated, but the usage boundary is unambiguous.

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