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

Goes well beyond the readOnlyHint annotation: states it is readable by any member like the REST read, that writes are out of lane, and flags the critical 'configured=false means every other field is the built-in default document, not a chosen setting' trap that must never be reported as a configured policy. It also warns that lowered_to_approval tools must never be described as running automatically. This is exactly the behavioral context annotations cannot carry.

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 tool's identity and the sibling disambiguation first, then details fields. It is a long, dense paragraph with no formatting, but given the absence of an output schema nearly every clause carries a semantic distinction an agent needs.

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?

No output schema exists, so the description must carry return-value meaning and it does: configured, autonomy_level enum, tier_overrides/auto_tier_tools/lowered_to_approval, cluster_scope modes, plus approvers, webhook id, action cap, cooldown and updated_at. Nothing needed to interpret a read 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?

The tool takes zero parameters, so the baseline is 4 and the description does not need to describe inputs. It instead documents returned field semantics (autonomy_level values, tier_overrides mapping, cluster_scope modes), which is output context rather than parameter meaning.

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+resource ('The organisation's AI auto-remediation policy') and identifies it as the read twin of a specific REST endpoint. It explicitly differentiates from siblings by naming get_ai_settings and pause_ai_autonomy and stating exactly what each does not cover, so an agent can select it without opening either 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?

Clearly positions the tool against named alternatives ('get_ai_settings reports only whether the policy is enabled, and pause_ai_autonomy is the org-wide soft switch, a different thing') and sets a boundary for writes ('writing the policy stays a portal or admin API lane'). It does not give an explicit positive trigger phrase like 'use when you need to inspect autonomy settings', but the context is unambiguous.

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