nuclei_customize

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

Annotations already declare readOnlyHint=true, idempotentHint=true, destructiveHint=false, so the safety profile is covered and the description's 'does not execute' clause is consistent with it. It adds only a modest amount beyond that: it clarifies this is a plan-editing step that neither runs scans nor approves them, but not whether base_plan is mutated or a new plan is returned, nor what happens on invalid flags.

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?

Three short sentences, front-loaded with the examples that carry most of the information, and no filler. It is slightly too compressed — the opening sentence lacks a subject/verb, which reads as a fragment rather than a stated purpose.

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 not be described. Still, for a tool that takes a nested base_plan plus a flag string, the description omits how the two interact (override vs. merge), which nested fields are reachable via flags, and whether invalid flags fail or are ignored — meaningful gaps for correct invocation.

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 0%, so the description must carry parameter meaning, and it does for the key 'parameters' string: the examples map flags (-rl, -c, -pt, -id, -severity, -timeout, -retries) onto schema fields like rate_limit, concurrency, protocols, template_ids, severities, request_timeout and retries, plus the crucial constraint 'Key=value is not supported'. The 'base_plan' parameter is left entirely unexplained, and the flag list is illustrative rather than exhaustive, keeping it short of a 5.

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 conveys the input format ('Accept CLI flags') and gives concrete examples, which lets an agent infer that it rewrites/mods a scan plan using nuclei-style flags. However, it never states a clear verb+resource (e.g. 'modify a scan plan'), so the action is inferred rather than declared and the distinction from nuclei_plan is only implicit.

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?

Negative exclusions ('Does not execute shell text or approve the plan') usefully rule out nuclei_execute and any approval step, which is real routing guidance. There is no positive when-to-use statement, no mention of when to prefer nuclei_plan or nuclei_recommend instead, and no note on what happens if flags conflict with base_plan.

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