run_security_benchmark

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

The description clearly frames the tool as a measurement/evaluation action rather than a data-mutating one, which is a useful behavioral signal. With no annotations, though, it does not disclose possible runtime cost, side effects, or what running a benchmark may do to the system.

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 a single compact sentence that front-loads the action and object. There is no filler, redundancy, or unnecessary detail.

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 zero-parameter tool, the description is mostly sufficient to decide whether to invoke it, and it clearly states what is being measured. However, there is no output schema and the description does not disclose what the benchmark returns, such as a report, pass/fail results, or metrics, which an agent would need to interpret the outcome.

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 has zero parameters, so there is no parameter-semantics burden on the description. The schema already covers 100% of the parameter surface, and the description appropriately treats this as a parameterless invocation.

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 names a specific verb ('Measure'), a specific resource ('Guardian correctness and decision latency'), and a context ('adversarial cases'), which clearly distinguishes this as a benchmark/evaluation tool. Even without a title, an agent can infer its role relative to the action/list sibling tools.

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 this is for evaluating Guardian on adversarial cases, and the tool name reinforces the benchmarking intent. However, it does not explicitly state when to prefer this over alternatives like evaluate_tool_call or guarded_action, and it offers no prerequisites or environment expectations.

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