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

Annotations indicate write access (readOnlyHint=false) and non-destructive hint, which aligns with the description's 'create or update' wording. The description adds behavioral context by explaining what each policy type does (e.g., approval_gate requires human approval), but it does not disclose whether updates replace or merge existing policies, leaving a transparency gap.

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 concise and well-structured: a clear opening statement followed by a bulleted list of the five policy types. Every sentence and bullet adds meaningful information, with no redundant content.

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?

The tool has no output schema, and the description does not mention return values or side effects. It also lacks details on how updates affect existing policies (replace vs. merge), which is important for a mutation tool. However, the parameter descriptions and policy type explanations cover much of the necessary context, making it adequate but not complete.

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 description coverage, the baseline is 3. The description goes beyond the schema by providing full semantics for the policyType enum values (e.g., brand_safety – 'Content must align with brand guidelines'), which helps agents construct valid policyType selections. It does not add detail for rules or active, but those are already adequately described in the schema.

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 action ('Create or update a policy/guardrail') and the target resource ('for an agent'), distinguishing it from read-only sibling tools like get_agent_policies. The enumeration of policy types with practical meanings adds specificity and confirms it is the tool for managing agent guardrails.

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 description implies when to use the tool—whenever an agent's policy needs to be created or updated—by listing the policy types that can be set. It does not explicitly mention alternatives or exclusions (e.g., 'to view policies use get_agent_policies'), but the context is clear given the sibling list includes a read-only counterpart.

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