creating-agents

A guide for creating separate coding agents: small AI work loops with their own model, allowed tools, and instruction file.

In plain words
What is it for?
Use it to add agents under an `agents/` directory, choose their model and tools, and define how each agent should work and what it returns.
Why use it?
It helps split off self-contained tasks so the main conversation stays focused and the task runs with controlled tools and instructions.

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/chinkan/rustfox/creating-agents
Any agent
npx skills add chinkan/RustFox --skill creating-agents
Clone the repo
git clone --depth 1 https://github.com/chinkan/RustFox

Made for: Claude Code, Codex.

Per session 35 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 879 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 $0.00035 $0.00879
Opus 5 $0.00017 $0.00439
Sonnet 5 $0.00007 $0.00176
Haiku 4.5 $0.00003 $0.00088

Measured 2d ago against content hash fad60694bec2, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

creating-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/creating-agents/SKILL.md · 106 lines

How it starts

The opening of the file, as written. The whole thing — 106 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Creating Agents

Agents are isolated agentic mini-loops with their own model, tool whitelist, and instruction file (AGENT.md). Use agents when you need to delegate a self-contained task to a separate LLM context — typically work that benefits from a different model, a restricted tool set, or clean isolation from the main conversation.

Skill vs Agent — When to Use Which

Use a skill (skills/) Use an agent (agents/)
Instruction to load into the main context Isolated task with its own model
No tool calls needed, or shares main tool set Needs a restricted tool whitelist
Workflow guidance or persona adjustment Sub-task that returns a single result
Lives in skills/<name>/SKILL.md Lives in agents/<name>/AGENT.md

Agent File Format

---
name: agent-name          # lowercase letters, numbers, hyphens only
description: One sentence — when to invoke this agent and what it returns.
model: provider/model-id  # required: the model this agent uses
tools: [tool1, tool2]     # required: exact tool names the agent may call
max_iterations: 5         # optional: default is global max (25)
tags: [tag1, tag2]        # optional
---

# Agent Name

(Instructions the agent follows. Starts with what it should do on first call.)

## Protocol

1. Step one
2. Step two
3. Return final answer

Tool Names

Tools must be the exact runtime names visible to the agent:

Category Name format Example
Built-in plain name read_file, write_file, execute_command
Skill tools plain name read_skill_file, write_skill_file, reload_skills
Agent tools plain name read_agent_file, write_agent_file, reload_agents
MCP tools mcp_{server}_{tool} mcp_google-workspace_query_gmail_emails

read_skill_file and read_agent_file are always available to every agent — no need to list them.

Read the full file on GitHub · 106 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 · 106 lines · 35 tokens per session scan A fad60694bec2

Subscribe to this mod's changes

creating-agents is a skill published in the GitHub repository chinkan/RustFox (7 stars, last pushed 26d ago), licensed MIT. It adds 35 tokens to every session and 879 once invoked, about $0.0002 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.