# Main Loop

> Use this skill for PRD-driven execution loops where one main agent orchestrates bounded sub-agents, keeps trying viable paths, logs what worked and what failed, and stops only when the PRD acceptance criteria pass or all PRD paths are exhausted.

- Skill: `bojak83318/main-loop` (Agent Skill)
- Install (CLI): `npx skillmds@latest add bojak83318/main-loop`
- Raw SKILL.md: https://api.skillmd.com/api/skills/bojak83318/main-loop/raw
- Safety review: pending
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: Product & Planning
- Author: bojak83318 (https://skillmd.com/u/bojak83318)
- Updated: 2026-09-22
- Page: https://skillmd.com/skills/bojak83318/main-loop

---


# Main Loop (Codex Wrapper)

**Activation Note:** This skill is a wrapper for the Codex skill in container `57db90a0b891`. To execute tasks, use `codex exec` non-interactively.

## Remote Execution Command
`incus exec kasm-vm -- docker exec 57db90a0b891 codex exec --skip-git-repo-check --dangerously-bypass-approvals-and-sandbox "[PROMPT]"`

## original Skill Content

Use this skill when the task is a long-running implementation or recovery workflow that must be driven by a PRD, not by ad hoc stopping points.