Summary

This skill provides a structured RFC-driven pipeline for decomposing large features into independently verifiable work units, coordinating multi-agent execution through DAG orchestration, quality gates, and merge queues. It guides agents through RFC intake, unit assignment, implementation, validation, and integration phases with explicit dependency management and rollback planning.

Detected Capabilities

Trigger Keywords

Phrases that MCP clients use to match this skill to user intent.

Risk Signals

Referenced Domains

External domains referenced in skill content, detected by static analysis.

Use Cases

• Break down large feature requests into parallel work units
• Coordinate multi-agent feature development with dependency tracking
• Establish quality gates and verification checkpoints for complex changes
• Manage integration risk when multiple teams work on interdependent features
• Implement staged rollout with dependency-aware merge ordering

Quality Notes

• Strong structural clarity: seven distinct pipeline stages are well-sequenced and logically ordered
• Unit spec template is concise but lacks detail on what constitutes valid `depends_on` lists, test coverage expectations, or risk_level definitions
• Complexity tiers are helpful scaffolding but would benefit from examples (e.g., 'adding a button = Tier 1, refactoring auth flow = Tier 3')
• Recovery section acknowledges stalled units but does not explain how to detect staleness or trigger recovery automation
• Missing: explicit error handling for dependency cycles, missing dependencies, or units that cannot be parallelized
• Outputs listed but not exemplified; agents may not know what 'integration risk summary' should contain
• Quality pipeline per unit is generic (research→tests→review) and does not reference domain-specific gates (e.g., security review for Tier 3)
• No guidance on how to handle schema changes or database migrations in multi-unit context