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

The description carries real behavioral weight the annotations do not: `run` is asynchronous with roughly an hour of latency, and `share` produces a link open to anyone signed in to the organization for 30 days, an access-control fact worth stating. Annotations only supply the generic readOnly/destructive/idempotent flags, so this added context is genuine value. Minor gap: it doesn't say whether `unshare` revokes links immediately or whether a new `run` replaces prior data.

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?

A single dense sentence front-loads the action enumeration, with each action's meaning in a parenthetical. It is efficient, though the telegraphic style (missing articles, semicolon-chained clauses) costs a little readability.

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-shape detail is unnecessary, and the description covers all four action modes plus the async and link-lifetime behaviors. What is missing is the default action and any note on prerequisites (e.g., whether connections must exist before `run`).

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 parameters, and it partly does: it defines what get/run/share/unshare mean and ties `run_id` to the share/unshare actions. It omits that `action` defaults to `get`, that `run_id` must be a positive integer, and whether get can target a specific historical run.

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?

Enumerates the four actions with their distinct effects: get (latest audit state or its report), run (a 30-day audit across connections), share/unshare (link management). A reader knows this is a multi-mode audit tool. It does not distinguish itself from audit-adjacent siblings like get_run/get_runs or access, so it falls short of a 5.

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?

The cost/timing note on `run` (30 days of data, ready within the hour) implicitly signals it is a slow, heavy action, and the share lifetime implies the sharing workflow. But there is no explicit when-to-use guidance, no mention that `action` defaults to `get`, and no routing to or away from siblings such as get_run.

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