What problem does it solve?
This Skill eliminates the guesswork and inconsistency of writing instructions for agents in a multi-agent orchestration plan, ensuring every agent receives clear, actionable guidance aligned with your project's harness state and goals.
Core Features & Use Cases
- Structured Prompt Generation: Creates per-agent, per-wave prompts formatted with standardized Goal/Context/Constraints/Done When fields for consistency across workflows.
- Goal-to-Chain Auto-Suggestion: Maps natural language task descriptions to recommended agent workflows, reducing the need to manually design orchestration plans from scratch.
- Built-In Quality Validation: Checks all generated prompts for achievable goals, measurable success criteria, and context alignment, flagging ambiguous or invalid instructions for your review before execution.
- Use Case: If you are running a multi-agent code review and reporting workflow, this Skill will generate tailored prompts for each assigned agent (e.g., code reviewer, report writer) with all necessary context and success criteria, eliminating the need to write each individual agent instruction manually.
Quick Start
Use the meta-prompt-builder skill to generate structured, validated prompts for each agent in your current agent-composer orchestration plan, including clear goals, context, constraints, and measurable done-when criteria for every wave.
