prompt-clarifying - Iteratively clarify and evaluate agent prompts
Iteratively improves agent-facing instructions through independent execution, two-sided evaluation, and targeted revisions until improvements plateau.
Tags
Updated: 2026-10-19Capabilities
Typical Inputs
Typical Outputs
What this skill does
- Clarify agent-facing instructions
- Prepare evaluation scenarios
- Delegate independent executions
- Evaluate execution results
- Track ambiguity and metrics
- Apply targeted prompt fixes
- Check convergence across iterations
Inputs
- Target agent instruction
- Official reference material
- Evaluation requirements
- Delegation mechanism
Outputs
- Revised agent instruction
- Evaluation scenarios
- Requirements checklist
- Executor reports
- Evaluation metrics
- Unclear-points analysis
- Failure-pattern ledger updates
Requirements
- Explicit user request
- Fresh independent executor
- Available delegation mechanism
- Access to relevant official references