decide_step

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

No annotations are provided, so the description carries the full burden. It explains inputs and output but does not disclose internal logic, potential side effects, or deterministic behavior. It hints at being a decision tool but lacks depth on how the decision is made or any constraints.

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 concise (about 100 Chinese characters) and packs purpose, input list, and output into a single sentence. It is front-loaded with the core goal. Minor improvement would be structuring with bullet points, but it is efficient for the content.

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 moderate complexity (5 parameters, no output schema), the description covers the tool's purpose, input semantics, and output nature. It lacks explicit return value structure (e.g., keys in response) but states '返回 mode 及理由', which is sufficient for basic use.

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 80%, meaning most parameters are already documented. The description adds context that inputs are 0-1 observables and briefly explains each parameter (e.g., '表面多关键'). This adds some value beyond the schema but does not significantly enhance parameter meaning.

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 clearly states the tool's purpose: to decide between System1 (intuition/cheap model) and System2 (strong model/deep reasoning) for the current step. It uses specific verbs ('判断', '该用') and identifies the resource ('当前这一步'). It is distinct from sibling tools like report_outcome or close_session, which handle other aspects.

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 explicitly lists the required inputs (criticality_hint, difficulty_hint, progress, context_pollution) and explains they are observable measures. It states the output (mode and reason) and clarifies this is a meta-cognitive decision orthogonal to picking steps. However, it does not provide explicit when-to-use or when-not-to-use guidance beyond the implied context.

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