create-agent

A guided scaffold for creating an AI agent focused on a specific area, such as code review or customer support. It creates the agent's identity, settings, skills, and workflows.

In plain words
What is it for?
Defining an agent's purpose, communication style, information sources, and multi-step work processes, then generating its starting structure.
Why use it?
It turns an idea for a specialist agent into the files needed to run and develop it.

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/gettalon/talon-plugins/create-agent
Any agent
npx skills add gettalon/talon-plugins --skill create-agent
Clone the repo
git clone --depth 1 https://github.com/gettalon/talon-plugins

Made for: Claude Code, Codex.

Per session 57 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 1,256 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 $0.00057 $0.01256
Opus 5 $0.00028 $0.00628
Sonnet 5 $0.00011 $0.00251
Haiku 4.5 $0.00006 $0.00126

Measured 2d ago against content hash 2ff7bcee9b88, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

create-agent 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.

plugins/agent-factory/skills/create-agent/SKILL.md · 132 lines

How it starts

The opening of the file, as written. The whole thing — 132 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Create Agent — Domain-Specific AI Agent Factory

Build a complete agent from scratch through guided brainstorming, then scaffold all files.

Process

Phase 1: Discovery

Ask these questions one at a time. Adapt based on answers.

  1. Domain: "What domain will this agent specialize in?" (e.g., financial research, code review, DevOps, customer support, legal analysis)

  2. Purpose: "What's the core job? What question does someone bring to this agent?" (e.g., "Is this stock undervalued?", "Is this PR safe to merge?")

  3. Personality: "How should this agent think and communicate?"

    • a) Analytical and precise (like a researcher)
    • b) Friendly and explanatory (like a tutor)
    • c) Terse and action-oriented (like an operator)
    • d) Opinionated with strong convictions
    • e) Custom — describe it
  4. Data Sources: "What information does this agent need access to?"

    • APIs (financial data, GitHub, Jira, etc.)
    • Web search
    • Local files / codebases
    • Databases
    • Browser automation
    • Custom tools
  5. Workflows: "What multi-step processes should this agent follow?"

    • e.g., "Gather data → Analyze → Validate → Present results"
    • e.g., "Read code → Find issues → Suggest fixes → Verify"
  6. Backend: "What LLM backend should it prefer?"

    • a) Default (whatever's configured)
    • b) Specific provider (glm, openrouter)
    • c) Local (ollama)
    • d) Multiple — route by task complexity

Phase 2: Design

Based on discovery, design these components:

Identity (SOUL.md):

  • Who is this agent? Give it a name and personality
  • What philosophy guides its decisions?
  • What does it value? What does it refuse to do?
  • Give the agent conviction and a clear point of view

Configuration (AGENT.yaml):

  • Model selection (sonnet for speed, opus for depth, haiku for lightweight)
  • Tool allowlist (restrict to what's needed)
  • Backend preference
  • Memory settings

Skills (SKILL.md files):

  • Each skill = one repeatable workflow the agent can execute
  • Design 2-4 core skills based on the domain
  • Each skill has: name, description, step-by-step checklist, output format
  • Each skill should have a clear trigger, step-by-step checklist, and structured output

Read the full file on GitHub · 132 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 · 132 lines · 57 tokens per session scan A 2ff7bcee9b88

Subscribe to this mod's changes

create-agent is a skill published in the GitHub repository gettalon/talon-plugins (6 stars, last pushed 5mo ago), licensed MIT. It adds 57 tokens to every session and 1,256 once invoked, about $0.0003 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.