assess_confidence

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

Describes return type (score with flags) and response to low confidence, but does not disclose side effects or state changes. As no annotations exist, description carries full burden but leaves scope for behavioral clarification (e.g., read-only, no persistence).

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?

Well-structured with trigger banner, purpose line, and labeled argument descriptions. Every sentence adds value, no redundancy. Front-loads the most critical usage instruction.

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?

Covers key aspects: purpose, trigger, output, and parameter semantics. Mentions uncertainty flags but does not detail them or the output schema. Given complexity (self-critique tool) and presence of an output schema, description is largely sufficient.

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?

Explains each parameter's role: claim ('proposed answer'), evidence ('supporting evidence/reasoning'), alternatives_considered ('other options rejected'), domain ('domain of expertise'). With 0% schema coverage, description adds meaning beyond raw parameter names.

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?

Clearly states verb ('assesses confidence'), resource ('important answer, recommendation, or architectural decision'), and output ('0-100 confidence score with specific uncertainty flags'). Distinguishes from siblings like 'calibration_score' and 'socratic_challenge' by specifying its role as a pre-delivery confidence check.

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?

Explicit trigger: 'Call this BEFORE delivering any important answer...' and guidance on low confidence ('recommends deeper analysis via sequential thinking or socratic challenge'). Mentions alternatives but does not compare to other sibling tools directly.

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