approval

Category: Data & AI | Uploader: sswym | Downloads: 0 | Version: v1.0（Latest）

Set omi approval posture (`suggest`, `auto`, `full-auto`) with explicit risk guardrails.

Changelog: Source: GitHub https://github.com/sswym/oh-my-iflow

Directory Structure

Current level: skills/approval/

📄 SKILL.md 865 B

SKILL.md

---
name : "approval"
description : "Set omi approval posture (`suggest`, `auto`, `full-auto`) with explicit risk guardrails."
---

## Purpose

Use this skill to control how aggressively omi runs autonomous actions versus asking for confirmation.

## Trigger

- User asks to reduce confirmation friction
- Autonomous mode is too conservative or too risky
- Team needs explicit approval policy before long runs

## Supported Modes

- `suggest`
- `auto`
- `full-auto`

## Workflow

1. Identify requested approval posture.
2. Explain side-effect and high-risk confirmation behavior for the selected mode.
3. Persist/update `.omi/state/approval.json` when filesystem tools are available.
4. Clarify that runtime approval flags remain host-CLI concerns.

## Output Template

```markdown
## Approval
- requested:
- applied:

## Guardrails
- ...

## Next Command
- ...
```

Comments 0

Latest | Hot

Please login before commenting.

No comments yet. Be the first one!

approval

Directory Structure

SKILL.md

Report

Notice