What problem does it solve?
Vague or incomplete task descriptions lead to misaligned execution, wasted effort, and rework in multi-agent development pipelines. This skill eliminates that risk by transforming raw task inputs into a clear, comprehensive, and human-validated brief that serves as the single source of truth for all downstream pipeline phases.
Core Features & Use Cases
- Context-Aware Brief Generation: Ingests existing project context (knowledge base files, AGENTS.md, README, optional prior research) to build a complete understanding of the task before asking any questions.
- Targeted Clarification: Only asks questions for gaps that cannot be inferred from available resources, avoiding unnecessary back-and-forth with the user.
- Structured Output & Validation: Produces a standardized brief covering task goal, scope, relevant files, architecture notes, quality gates, and open questions, then requires explicit human validation before allowing the pipeline to proceed.
- Use Case: For a team using the Roster pipeline, assigning a "add user authentication" task will generate a validated brief outlining in-scope components, relevant auth-related files, exact test commands, and open questions about OAuth provider support, so the implementation phase has zero ambiguity.
Quick Start
Invoke the roster-intake skill with your raw task description to generate a validated, unambiguous brief that the rest of the Roster pipeline will use as its single source of truth.