LogoClawIndex
CasesSkillsAbout
LogoClawIndex

loop-factory - Spec-driven agent workflow and execution loop

Runs a spec-driven agent loop to implement, verify, review, and archive task specifications using Claude Code or Codex CLI.

Tags

Updated: 2026-09-18
agent-loopspec-driventask-automationclaude-codecodexcli

Capabilities

Scan task specs in queueDispatch specs to AI agentsImplement code based on specsRun verification commands

Typical Inputs

Markdown spec filesVerification commandsGit repository

Typical Outputs

Prompt files in factory/prompts/Run records in factory/runs/Review files in factory/reviews/

What this skill does

  • Scan task specs in queue
  • Dispatch specs to AI agents
  • Implement code based on specs
  • Run verification commands
  • Review code changes against criteria
  • Archive passed task specs
  • Backpropagate system changes to docs
  • Scaffold factory folders in repos

Inputs

  • Markdown spec files
  • Verification commands
  • Git repository
  • Claude Code or Codex CLI

Outputs

  • Prompt files in factory/prompts/
  • Run records in factory/runs/
  • Review files in factory/reviews/
  • Archived spec files in factory/specs/archive/
  • Updated source code and documentation

Requirements

  • Python 3.10+
  • Git environment
  • Claude Code CLI or Codex CLI

Source

  • Spec: SKILL.md

ClawIndex

OpenClaw Skills & Use Case Index

ClawIndex is an ecosystem-driven index of OpenClaw skills and real-world use cases.

Index

Skills·
Cases

Meta

About·
Disclaimer·
Email·
GitHub
© 2026 ClawIndex All Rights Reserved.