# Agent Cli Delegation

> Umbrella workflow for delegating coding work to local agent CLIs such as Claude Code, Codex, and OpenCode.

- **Type:** Skill
- **Install:** `agentstack add skill-s3yed-appie-kit-agent-cli-delegation`
- **Verified:** Yes — security-reviewed for prompt injection and unsafe behavior
- **Seller:** [S3YED](https://agentstack.voostack.com/s/s3yed)
- **Installs:** 0
- **Category:** [AI & ML](https://agentstack.voostack.com/c/ai-and-ml)
- **Latest version:** 0.1.0
- **License:** MIT
- **Upstream author:** [S3YED](https://github.com/S3YED)
- **Source:** https://github.com/S3YED/appie-kit/tree/master/skills/autonomous-ai-agents/agent-cli-delegation
- **Website:** https://weblyfe.ai

## Install

```sh
agentstack add skill-s3yed-appie-kit-agent-cli-delegation
```

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

## About

# Agent CLI Delegation

Use this umbrella when the user wants coding work delegated to a local autonomous coding CLI rather than completed entirely in the current Hermes turn.

## Choose the agent CLI

| CLI | Best fit | Notes |
|---|---|---|
| Claude Code | Larger refactors, repo-aware implementation, PR-ready changes | Keep prompts scoped and include verification commands. |
| Codex | OpenAI Codex CLI feature/fix work and iterative patching | Provide exact paths, constraints, and tests to run. |
| OpenCode | Alternative coding agent and PR-review workflows | Useful when the user prefers OpenCode or when comparing agents. |

## Delegation prompt shape

Include:

1. Goal and non-goals.
2. Repository path and relevant files.
3. Constraints: style, dependencies, APIs, security, backwards compatibility.
4. Required verification commands.
5. Expected deliverables: patch, tests, summary, risks.

## Execution rules

- Run agent CLIs in a tracked terminal/background process, not an untracked shell background.
- Require the child agent to report changed files and verification output.
- Independently verify critical claims: inspect diffs and run tests/checks yourself.
- Never ask a child agent to use credentials, deploy, merge, or delete without explicit user authorization.

## Review after delegation

1. `git status --short` and `git diff`.
2. Check for unrelated changes, secrets, generated junk, or destructive edits.
3. Run the test/lint/typecheck commands that matter.
4. Summarize what was done and what remains.

## Source & license

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

- **Author:** [S3YED](https://github.com/S3YED)
- **Source:** [S3YED/appie-kit](https://github.com/S3YED/appie-kit)
- **License:** MIT
- **Homepage:** https://weblyfe.ai

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-s3yed-appie-kit-agent-cli-delegation
- Seller: https://agentstack.voostack.com/s/s3yed
- 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%.