Agent Cli Delegation

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
• Source: S3YED/appie-kit
• License: MIT
• Homepage: https://weblyfe.ai

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