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

Annotations already declare destructiveHint=true and idempotentHint=true; the description adds real context on top: the required write scope, the human-confirmation gate, the guarantee that the finding, evidence and score are preserved, and the meaning of the intent_saved_refresh_pending conflict plus why repeating the call is pointless. This is exactly the behavioral detail annotations cannot express.

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 scope, then prerequisites, then the conflict semantics. Dense but every sentence carries operational information; the only mild drag is the compound final sentence about the linked action refreshing on reconciliation.

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, four-parameter mutation with no output schema and no schema descriptions, the description covers authorization, human confirmation, state preservation, and the primary known conflict reason. It omits what the call actually changes (removes vs. modifies the restriction) and what a successful return conveys, but nothing critical to invoking it correctly 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?

With 0% schema description coverage, the description must carry the parameter burden. It adds genuine meaning for urls ('only the exact URLs displayed in an audit finding') and confirmedIntent ('explicit customer confirmation', 'never infer customer intent'), and implies auditRunId/findingId by tying everything to an audit finding. However it never mentions the 1-4 URL cap, the UUID format of auditRunId, or that confirmedIntent must literally be true, leaving gaps the schema cannot fill.

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 ('Reconsider') and resource ('an intentional retrieval restriction') and narrows scope to 'only the exact URLs displayed in an audit finding'. That scope constraint makes it distinguishable from a general restriction/acknowledgement operation without needing to open a 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 explicit preconditions ('Requires explicit customer confirmation and actions.write'), an exclusion ('Never infer customer intent'), and a concrete do-not-call condition after an intent_saved_refresh_pending conflict. It stops short of naming the sibling tool (e.g. acknowledge_retrieval_restriction) as the alternative path, so routing between the two still requires inference.

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