# Loop

> Repeat audit then 2080 review; reuses models config. Manual invoke only.

- Skill: `avisheku/loop` (Agent Skill, multi-file: 5 files)
- Install (CLI): `npx skillmds@latest add avisheku/loop`
- Raw SKILL.md: https://api.skillmd.com/api/skills/avisheku/loop/raw
- Safety review: pending
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: Security
- Author: avisheku (https://skillmd.com/u/avisheku)
- Updated: 2026-09-21
- Page: https://skillmd.com/skills/avisheku/loop

---


# /loop

Periodic improvement loop — **manual** via script (no daemon).

```powershell
.\scripts\Invoke-Loop.ps1
.\scripts\Invoke-Loop.ps1 -Iterations 2
```

Config: `config/loop.json` — steps `audit`, `2080`.

Reuses: `Invoke-AuditReport`, `/2080` roles, model tips path.

Enable `loop.json` `enabled: true` only when you want repeated iterations.