# Sd Loop

> Start an autonomous loop — re-feeds the same prompt until done

- **Type:** Skill
- **Install:** `agentstack add skill-stevederico-skills-sd-loop`
- **Verified:** Yes — security-reviewed for prompt injection and unsafe behavior
- **Seller:** [stevederico](https://agentstack.voostack.com/s/stevederico)
- **Installs:** 0
- **Category:** [Agent Skills](https://agentstack.voostack.com/c/agent-skills)
- **Latest version:** 0.1.0
- **License:** MIT
- **Upstream author:** [stevederico](https://github.com/stevederico)
- **Source:** https://github.com/stevederico/skills/tree/master/sd-loop

## Install

```sh
agentstack add skill-stevederico-skills-sd-loop
```

Requires the [AgentStack CLI](https://agentstack.voostack.com/docs/cli). Works with Claude Code, Cursor, and any MCP-compatible agent.

## About

# Start Loop

Parse the user's arguments: `/sd-loop [""] [--max-iterations N] [--completion-promise "TEXT"]`

Defaults:
- prompt (if none provided): `Read SPEC.md and implement everything in it. When all items are complete, commit your work, push the branch, and submit a PR with gh pr create. Then output DONE.`
- max-iterations = 50
- completion-promise = "DONE" (when using default prompt), otherwise none

## Instructions

1. If using the default prompt (no custom prompt provided), check if `SPEC.md` exists in the current project directory. If not, create it with this starter template and tell the user to fill it in before running `/sd-loop` again — then stop:

```markdown
# Spec

## Requirements

-
```

2. Create `.claude/loop.local.md` in the current project directory:

```markdown
---
active: true
iteration: 0
max_iterations: 
completion_promise: ""
started_at: 
---

```

3. Confirm to the user:
   - Loop started with their prompt
   - Max iterations: N
   - Completion promise: TEXT (if set) — output `TEXT` to exit early
   - To cancel: `/sd-loop-cancel`

4. Then immediately execute the prompt — begin working on it now.

## Source & license

This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.

- **Author:** [stevederico](https://github.com/stevederico)
- **Source:** [stevederico/skills](https://github.com/stevederico/skills)
- **License:** MIT

Install and usage instructions live in the source repository linked above.

## Pricing

- **Free** — Free

## Security capabilities

Automated source analysis of v0.1.0 — what this tool can access:

- **Network access:** no
- **Filesystem access:** no
- **Shell / process execution:** no
- **Environment & secrets:** no
- **Dynamic code execution:** no

*"Yes" means the capability is present in the source — more access means more to trust, not that it is unsafe.*


## Versions

- **0.1.0** — security scan: passed — Imported from the upstream source.

## Links

- Listing page: https://agentstack.voostack.com/l/skill-stevederico-skills-sd-loop
- Seller: https://agentstack.voostack.com/s/stevederico
- Browse the marketplace: https://agentstack.voostack.com/browse

---
Listed on AgentStack — the marketplace for AI agent skills and MCP servers. Every listing is security-reviewed. Creators keep 70%.