refine_threat_model

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

With no annotations present, the description provides extensive behavioral details: it creates a new version, reports progress, prevents silent replacement/dropping of entities, handles semantic rejections via 'semantic_rejections', ensures stable CO IDs, and uses soft-delete via remove_entity. This is excellent transparency for a complex mutation tool.

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 structured with bullet points and sections, making it easy to scan. It is longer than average but every detail serves to clarify critical behavior (preservation, rejection, versioning). Could be trimmed slightly but it's well-organized.

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?

Given the complexity of the tool (semantic preservation, rejection arrays, versioning, tombstones), the description covers all critical aspects. The presence of an output schema for return values reduces the need to describe them. This is complete enough for an agent to use correctly.

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 covers 67% of parameters (model_id, instruction). The description adds context for 'instruction' via example but does not explain server_version. Since schema coverage is moderate, the description provides minimal additional value beyond the schema. Baselines: with >80% coverage, score is 3, but here it's 67% so expect more, but description doesn't add much about params, so 3 is fair.

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 first sentence clearly states the action ('Refine an existing threat model based on an instruction') and specifies what is refined (assets, attackers, trust boundaries, control objectives). This distinctly separates it from siblings like generate_threat_model or edit_asset, as it's for whole-model refinement via natural language instruction.

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 communicates when to use this tool (to refine an existing model via instruction) and implicitly contrasts with other tools by describing what it cannot do (e.g., silently replace entities) and referencing remove_entity as the way to delete entities. However, it does not explicitly compare to alternative tools like edit_asset or generate_threat_model, so there's room for explicit differentiation.

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