directory_audits

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

With no annotations, the description carries the full burden. It discloses read-only status, required permissions (AuditLog.Read.All, Reports Reader), retention (30 days), and critically explains client-side filtering for targetResources—including the consequence for max_pages budgeting and the capped=true warning. This is thorough and actionable.

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?

The description is longer than typical but every sentence earns its place—purpose, example, technical caveat, permissions, retention, and parameter docs. It's front-loaded with the central question, and the structure (overview, rationale, behavior, args) is logical and scannable. Slight verbosity in the anecdote, but not wasteful.

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 tool with 4 parameters, no output schema, and no annotations, the description covers purpose, usage, permissions, pagination behavior, retention, and parameter semantics. It also anticipates edge cases (capped=true, larger max_pages). The only minor gap is an explicit return format, but the description already implies the audit fields (initiated_by, target_resources).

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%, but the Args section explains each parameter's meaning, default, and clamp ranges (hours [1,720], top [1,500], max_pages budget). It also clarifies the semantic nuance of 'user' (actor or target). This fully compensates for the schema's lack of descriptions.

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 clearly states the resource (directory audits), the core function (who did what to the directory), and distinguishes it from signin_logs as the operator-side counterpart. It even provides a concrete scenario (manual unblock and reset) that clarifies its unique role. This fully differentiates it from siblings.

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?

Explicitly ties usage to a specific triage need (distinguishing 'handled by human' from 'still open'), names the alternative (signin_logs) and contrasts behavior (full-window fetch vs server-side filtering). It also gives operational advice about max_pages requirements, making when-to-use and when-not-to-use precise.

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