context-driven-development

Context-Driven Development (CDD) is a methodology that treats project documentation as a first-class, version-controlled artifact. This skill automates the creation and maintenance of a 'conductor/' directory, serving as the single source of truth for AI agents and human teams. By establishing persistent, structured foundations, it ensures that AI interactions remain consistent throughout the project lifecycle. It is ideal for software engineers, project leads, and AI-assisted development teams who need to reduce ad-hoc prompting and maintain long-term alignment across complex, evolving codebases.

• Automatically scaffolds and maintains core artifacts: product.md (vision/goals), tech-stack.md (dependencies/architecture), workflow.md (process/standards), and tracks.md (work management).

• Facilitates greenfield setup by interactively defining project standards, and brownfield onboarding by extracting context from existing code, configurations, and documentation.

• Enforces a clear Context-Spec-Plan-Implement workflow, ensuring specifications and task plans are validated before code generation begins.

• Provides seamless synchronization, ensuring that feature completions or dependency changes are reflected across all relevant documentation.

• Enables semantic reversion capabilities by linking git commits to track-based work units.

• Always run /conductor:setup to initialize or detect project patterns.

• Use the Context-Driven Development workflow to prevent scope drift and maintain architectural integrity.

• Regularly update product.md when features are completed or roadmaps shift to keep the AI aligned with current business intent.

• Verify context artifacts before starting any new development track to flag outdated information.

• Manage tech-stack.md strictly; document the rationale before adding new dependencies to prevent tech debt.