external-coding-agents

A guide for delegating coding work to command-line AI agents such as Claude Code, Codex, and OpenCode. These agents can inspect files, change code, run shell commands, and work with Git repositories.

In plain words
What is it for?
Use it to delegate refactors, bug fixes, CI/CD automation, pull-request reviews, and repository research, then inspect the changes and verification results.
Why use it?
It helps decide when delegation is useful and gives a consistent way to describe the task and required checks. This is especially useful for multi-file changes, automated pipelines, code reviews, and independent investigation.

Skill for Claude CodeCodex

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add skills/s3yed/appie-kit/external-coding-agents
Any agent
npx skills add S3YED/appie-kit --skill external-coding-agents
Clone the repo
git clone --depth 1 https://github.com/S3YED/appie-kit

Made for: Claude Code, Codex.

Per session 26 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 1,080 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5 $0.00026 $0.01080
Opus 5 $0.00013 $0.00540
Sonnet 5 $0.00005 $0.00216
Haiku 4.5 $0.00003 $0.00108

Measured 2d ago against content hash 213aeb53d815, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

external-coding-agents scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

skills/agentic/external-coding-agents/SKILL.md · 108 lines

How it starts

The opening of the file, as written. The whole thing — 108 lines — stays where its author put it; the contents beside it link to each section on GitHub.

External AI Coding Agents

Delegate coding work to third-party AI coding agents from within Hermes. These tools (Claude Code, Codex, OpenCode) are autonomous CLI agents that can read files, write code, run shell commands, and manage git workflows — each with its own strengths.

When to Delegate vs Do It Yourself

Situation Approach
Bug fix in a known file Do it yourself with patch/write_file — faster, no overhead
Complex multi-file refactor Delegate — the external agent has persistent context across files
Task needs a fresh perspective Delegate — the external agent will spot things you might miss
CI/CD automation Delegate — use print mode for clean automation
PR review of someone else's code Delegate — the external agent can see the full diff in context
Research/reconnaissance Delegate — agent can browse and explore independently
Simple one-line change Do it yourself

Common Orchestration Patterns

All three coding agents use the same core patterns:

Print Mode (Non-Interactive — Preferred)

Cleanest integration. No PTY, no dialog handling, structured output:

terminal(command="<agent> -p 'Add error handling to all API calls in src/'", workdir="/path/to/project", timeout=120)

PTY Background Mode (Interactive)

For iterative work needing multi-turn conversation. Use background=true, pty=true and process tool to monitor:

terminal(command="<agent>", workdir="~/project", background=true, pty=true)
process(action="submit", session_id="<id>", data="Implement OAuth refresh flow")
process(action="poll", session_id="<id>")
process(action="log", session_id="<id>")
process(action="kill", session_id="<id>")

Get the Diff Back

After delegation completes, always inspect what changed:

git diff
git diff --stat

Agent Decision Guide

Need Best Agent Why
Best overall coding ability Claude Code Anthropic's most capable model, v2.2+, rich feature set
Cheapest / open-source preference OpenCode Provider-agnostic (OpenRouter, Anthropic, OpenAI, local models)
OpenAI ecosystem Codex OpenAI's autonomous agent, OAuth integration
PR review with inline comments Claude Code Built-in --from-pr flag, formal review workflow
Batch multi-issue fixing Codex or OpenCode Lightweight, parallel worktree pattern
Long-running autonomous session Any Use background=true + process tool for monitoring

Read the full file on GitHub · 108 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 2d ago First seen · 108 lines · 26 tokens per session scan A 213aeb53d815

Subscribe to this mod's changes

external-coding-agents is a skill published in the GitHub repository S3YED/appie-kit (6 stars, last pushed 6d ago), licensed MIT. It adds 26 tokens to every session and 1,080 once invoked, about $0.0001 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.