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

Annotations already declare destructiveHint=true, idempotentHint=true and openWorldHint=false, so the safety profile is partly covered. The description adds genuinely non-structured context: findings are stored as "agent-reported, not verified deployed behavior", a group label does not merge reports, and it returns the updated audit decision plus report link. It still does not say what a write overwrites or how a mismatched revision is handled.

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-loads the purpose and keeps three short sentences, with the trust caveat ("agent-reported, not verified") placed early where it matters. The first sentence is a dense comma-run listing many data categories, but for a 19-parameter ingestion tool that enumeration earns its length.

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?

An output schema exists, so return-value detail is not required, yet the description still tells the agent it gets back an updated audit decision and report link. Combined with the trust model, group semantics and unknown-handling rules, an agent has enough to call a complex nested-object tool correctly; only the write/overwrite and revision-conflict behavior remain unstated.

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 100% with 19 well-documented properties, many with their own enums and per-field notes, so the schema carries the parameter burden and baseline 3 applies. The description's parameter-level remarks (group is label-only, unknown != failed) largely restate what the schema already says ("legacy grouping label only; does not combine reports"), adding little beyond it.

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?

Opens with a specific verb+resource pair ("Saves code review findings ... to the audit") and enumerates exactly which findings are captured (server-side events, dedup IDs, Pixel/container presence, payment provider, framework, backend location, existing CAPI quality, reviewed paths, N/A checks). This distinguishes it from read-side siblings like read_audit, though it does not explicitly contrast with the other write-side siblings such as record_audit_context or record_audit_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?

Usage context is implied through field-level guidance ("Unknown values stay unknown, not failed", "Missing access counts as unknown, not N/A") which tells the agent how to fill the report, but there is no statement of when to call this versus begin_audit, record_audit_context, or save_audit_to_project. No prerequisites about needing audit_id/audit_key/revision are called out as a gating sequence.

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