# AI Automation Workflows

> Use when you need to orchestrate an approved AI workflow, scheduled automation, or execution backend without bypassing the packet and approval path. This local mirror aligns OpenClaw with inference-sh agent-skills ai-automation-workflows.

- Skill: `sheshiyer/ai-automation-workflows` (Agent Skill)
- Install (CLI): `npx skillmds@latest add sheshiyer/ai-automation-workflows`
- Raw SKILL.md: https://api.skillmd.com/api/skills/sheshiyer/ai-automation-workflows/raw
- Safety review: pending
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: AI & ML
- Author: Sheshiyer (https://skillmd.com/u/sheshiyer)
- Updated: 2026-09-17
- Page: https://skillmd.com/skills/sheshiyer/ai-automation-workflows

---


# AI Automation Workflows

Use this skill when the main problem is workflow choreography, not authorship.

## Primary Owner

- `sadhana-orchestrator`

## OpenClaw Bindings

- Execute approved inference tasks:

```bash
python3 ~/.openclaw/scripts/inference_flow_run.py --packet /abs/path/to/packet.json --dry-run
```

- Hand actionable findings to Paperclip:

```bash
python3 ~/.openclaw/scripts/handoff-to-paperclip.py --packet /abs/path/to/packet.json
```

- Preflight the distribution lane:

```bash
python3 ~/.openclaw/scripts/test-distribution-flow.py
```

## Workflow

1. Confirm that the upstream packet, approval state, and required assets exist.
2. Confirm whether the run is `dry-run` or live.
3. Route execution through the existing adapters.
4. Capture receipts and surface failures.
5. Never skip the approval boundary just because automation is available.

## Guardrails

- This skill does not create permission to publish.
- Cron-owned automation must remain downstream of `CLIP` and the packet gate.
- Prefer existing adapters over raw provider calls.