S
Spec-Driven Implement
Implementation system driven by task specs with LLM-as-Judge auto-verification, iterative repair, breakpoint resume, and human-in-the-loop checkpoints
ByNeoLabHQ
WhatIsIt
- Read the task file ONCE (Phase 1 only)
- Launch sub-agents via Task tool
- Receive reports from sub-agents
- Mark stages complete after judge confirmation
- Aggregate results and report to user
HowToUse
Install this skill in your Claude environment to enhance spec-driven implement capabilities. Once installed, Claude will automatically apply the skill's guidelines when relevant tasks are detected. You can also explicitly invoke it by referencing its name in your prompts.
The full source and documentation is available on GitHub.
KeyFeatures
- Implementation system driven by task specs with LLM-as-Judge auto-verification, iterative repair, breakpoint resume, and human-in-the-loop checkpoints
- Automated workflow execution for repetitive tasks
- Seamless integration with Claude's development workflow