audit

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

Annotations already establish readOnlyHint=true and openWorldHint=false, so the safety profile is covered. The description adds genuinely useful behavior beyond that: the specific categories of findings and the fact that each finding carries a suggested fix, which tells the agent the output is actionable rather than just a list. It does not discuss cost, scope limits, or how the stale threshold interacts with the scan.

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?

Two sentences, front-loaded with the core purpose, then a compact enumeration of finding types and the 'each finding suggests a fix' payoff. No filler, no repetition of the tool name.

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 values need not be explained, and the description still conveys the actionable nature of results. The one remaining gap is routing guidance among many sibling diagnostic tools, which for a 45-sibling surface is a meaningful omission.

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 100% schema coverage and a single optional parameter already described in the schema ('Flag open theses unchanged this long', default 90, range 7–3650), the schema carries the parameter burden. The description adds no further meaning about stale_days or how it affects which findings appear.

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 names a specific operation ('health check of the research graph') and then enumerates the exact defect classes it surfaces: expired/thinly sourced assumptions, past-horizon or stale theses, conflicting claims, unconfirmed starter links, orphans. That is far more concrete than the bare name 'audit', though it never distinguishes itself from closely related diagnostics like scorecard, thesis_weak_points, or digest.

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 is only implied by the phrase 'health check' – an agent can infer this is a diagnostic sweep, but nothing states when to run it versus scorecard, thesis_weak_points, list_alerts, or digest, and there are no exclusions or prerequisites.

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