nuclei_select_targets

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

Annotations already declare readOnly, idempotent, non-destructive, closed-world, and the description is consistent with them while adding genuine behavior: missing decisions keep candidates by default, and shared IP/title must not merge virtual hosts or cross protocol/port boundaries. Those default and merge rules are exactly the kind of detail annotations cannot express. Return format is not described, but an output schema exists.

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?

It is short, but written in telegraphic fragments ('First omit decisions to read numbered candidates') whose meaning is hard to parse on first read, and the exclusion sentence is front-loaded as a run-on. The information is there but the structure is choppy rather than cleanly sequenced.

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 two-phase workflow tool with an output schema, the decision loop and merge constraints are covered, so an agent can roughly drive it. The gap is input-side: with 0% schema coverage and no explanation of intent/targets/assets, the agent cannot confidently construct the first call, which is a material omission for a required-parameter tool.

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 description coverage is 0%, so the description carries the full burden. It does document the decisions element shape ({id,action,reason,duplicate_of}) and the keep/exclude/duplicate action vocabulary, which is valuable. However, the required 'intent' and 'targets' parameters and the optional 'assets' parameter are never explained – what format targets take or what intent governs is left to guesswork.

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 opens with 'Prepare or validate host-AI target selection', a specific verb pair, but 'host-AI target selection' is undefined jargon and the two-phase nature (propose candidates, then submit decisions) only becomes clear from the later sentences. The negative clause ('No model/API calls, probing, approval or execution') does help separate it from execution siblings like nuclei_execute, but no sibling is named directly.

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?

It gives a concrete ordered workflow: first call without decisions to read numbered candidates, return decisions, present the audit, then approve an exact plan. That is real when-and-how guidance. What is missing is routing against the closest siblings (nuclei_plan, nuclei_recommend, nuclei_execute) – the agent must infer that this tool never executes.

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