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

Beyond the annotations (readOnly, idempotent, non-destructive, closed-world), the description discloses the return shape (every attack surface found, ranked by severity, with evidence spans), the exact cost ($1.50 USDC), the payment mechanics (EIP-3009 transferWithAuthorization in an X-PAYMENT header, retry-on-402), and the failure behavior of an unpaid call. That is substantive behavioral context the annotations cannot convey.

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?

Purpose is front-loaded in the first sentence, and the payment block is arguably necessary for a paid endpoint. The VALUE section restates some of what the first sentence already implies, and the SDK plug is marginal, so the text runs longer than needed but nothing is confusing.

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 no further explanation, and payment/auth details are fully covered. The remaining gap is that all six parameters are optional with defaults, yet the description never says whether at least one input group must be supplied or what an empty call yields.

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 coverage is 100%, but the schema 'descriptions' are only endpoint tags such as '[POST /v1/security/prompt-injection]', so the description carries real semantic weight by explaining that the prompt.* group scans prompts, tool descriptions and tool results, while tools.* scans tool definitions and credentials.* scans text. It never mentions the credentials.files parameter, leaving that one input's expected content unstated.

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 and resource ('Assess the threats TO an agent') and enumerates the three detectors it composes: prompt-injection over prompts/tool descriptions/tool results, tool-poisoning over tool definitions, and credential detection over text. This distinguishes it from single-surface siblings like scan_prompt_injection and audit_tool_poisoning, though those tools are never named explicitly, so the agent must infer the distinction.

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 VALUE paragraph gives a real selection rule: the three surfaces are assessed together because 'assessing one of them alone produces a clean report that means little.' That tells the agent when this composite is preferable to a single detector. It stops short of naming the narrower siblings or stating exclusions (e.g. when a single scan is sufficient).

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