# Execute

> Use before beginning any multi-step task with multiple files, tests, or verification steps.

- **Type:** Skill
- **Install:** `agentstack add skill-mverab-slashstack-skills-execute`
- **Verified:** Yes — security-reviewed for prompt injection and unsafe behavior
- **Seller:** [mverab](https://agentstack.voostack.com/s/mverab)
- **Installs:** 0
- **Category:** [AI & ML](https://agentstack.voostack.com/c/ai-and-ml)
- **Latest version:** 0.1.0
- **License:** MIT
- **Upstream author:** [mverab](https://github.com/mverab)
- **Source:** https://github.com/mverab/slashstack-skills/tree/main/skills/execute
- **Website:** https://slashstack.dev

## Install

```sh
agentstack add skill-mverab-slashstack-skills-execute
```

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

## About

# Execute

## Purpose

Turn multi-step tasks into goals with explicit scope, acceptance criteria,
verification commands, and completion expectations.

## When to Use

Use before beginning any multi-step task that involves multiple files, tests,
or verification steps.

## Workflow

1. Read the registry: open `.agents/memory/goal-registry.json` to see active and
   completed goals.
2. Define scope: list the file path globs this goal may read or write.
3. Set acceptance criteria: describe what "done" looks like in a single sentence.
4. List verification commands: the test, build, or lint commands that must pass.
5. Declare commit and push expectations: whether the goal requires a commit
   and/or a push upon completion.
6. Check for scope overlaps: if any active goal's scope intersects with yours,
   report the overlap, name the conflicting goal id, and refuse to start until
   the conflict is resolved.
7. Do the work: edit only files within the declared scope.
8. Run verification: execute every command in the verification list and record
   the pass/fail result.
9. Mark complete with evidence: update the goal `status` to `completed`, set
   `updatedAt`, and record a brief evidence summary.
10. Archive stale goals: set `archived: true` on completed goals older than 30 days.

## Output

If complete: updated goal entry with `status: completed`, `updatedAt`, and
an evidence summary.

If abandoned: updated goal entry with `status: abandoned`, `updatedAt`, and
a reason.

## Source & license

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

- **Author:** [mverab](https://github.com/mverab)
- **Source:** [mverab/slashstack-skills](https://github.com/mverab/slashstack-skills)
- **License:** MIT
- **Homepage:** https://slashstack.dev

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-mverab-slashstack-skills-execute
- Seller: https://agentstack.voostack.com/s/mverab
- 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%.