Summary

Lean Build is a high-level architecture and development methodology skill that guides AI agents to implement feature work by deriving acceptance criteria, tracing system layers, delivering coherent end-to-end paths, and reusing existing seams. It emphasizes constraint-driven development, strict scope boundaries, and explicit stop conditions to prevent overbuilding and maintain architectural coherence.

Detected Capabilities

Trigger Keywords

Phrases that agents use to match this skill to user intent.

Use Cases

• Implement new product features with bounded scope and clear acceptance criteria
• Design feature slices that fit existing system architecture without speculative extensions
• Refactor duplicated behavior by identifying and reusing appropriate seams
• Trace feature requirements through layers to ensure each component has clear ownership
• Stop implementation at acceptance rather than overbuilding with unnecessary polish or configuration

Quality Notes

• Skill content is highly abstract and domain-specific (appears tailored to a specific organization's 'Native Core' architecture)
• Instructions prioritize architectural principles over concrete, step-by-step procedural guidance
• Language is prescriptive but lacks concrete examples of what successful implementation looks like
• No explicit error handling or edge case guidance provided
• Scope is well-bounded (architectural compliance, acceptance-driven stop condition) but very implicit
• Unclear how an agent would validate adherence to principles like 'never force work into one file' or 'reuse fitting seam' without domain knowledge
• Support file (agents/openai.yaml) provides a default prompt that contextualizes the skill but does not add procedural detail