Experienced engineering flow with high-fidelity, deterministic results.
dev.kit resolves the Drift (intent divergence) by Normalizing it into a deterministic path and Iterating to resolution. It acts as a Thin Empowerment Layer that bridges human intent with repository truth via Context-Driven Engineering (CDE).
Intent is mapped to repo-specific logic and normalized into a deterministic execution plan.
Capabilities are resolved at runtime through a Dynamic Discovery Engine that scans script headers and skills.
AI iterates through instruction steps with deterministic validation and real-time status updates.
dev.kit provides a Thin Empowerment Layer that bridges chaotic intent with deterministic repository functions.
| Phase | Description | Implementation |
|---|---|---|
| 1. Grounding | UDX Worker provides the isolated, pre-hydrated base environment. | udx/worker |
| 2. Discovery | Skill Mesh resolves internal commands and @udx NPM packages. |
@udx/mcurl |
| 3. Normalize | AI Reasoning Skills transform intent into a bounded workflow.md. |
SKILL.md |
| 4. Execute | Deterministic Primitives execute the logic within the CLI boundary. | lib/commands/ |
dev.kit status: (Default) High-fidelity engineering brief and system health.dev.kit doctor: Verify environment health, software detection, and compliance.dev.kit ai: Unified agent integration management and grounding.dev.kit sync: Logical, atomic repository synchronization and drift resolution.dev.kit task: Manage the lifecycle of active workflows and engineering sessions.
Execution: Run any repository-bound skill with
dev.kit skills run <name>.
The dev.kit knowledge base is structured to reflect CDE Principles.
- Foundations: Core philosophy (CDE), dev.kit primitives, and methodology.
- Runtime: CLI overview, lifecycle, and execution loops.
- AI Integration: Grounded orchestration and agent mission.
- Best Practices: High-fidelity engineering rules and command mappings.
curl -fsSL https://udx.dev/dev.kit/install.sh | bashThe dev.kit mission is grounded in foundational research on high-fidelity automation and AI orchestration:
- The Power of Automation: Systematic transformation of the engineering flow.
- Observation-Driven Management: Revolutionary task normalization through pattern recognition.
- Autonomous Technical Operations: Principles for high-fidelity agent grounding and execution.
UDX DevSecOps Team