FoundationsAgent systems topic

Skills

Skills package reusable instructions, examples, scripts, and resources for a repeatable workflow. They should stay focused enough to load only when relevant and concrete enough that another agent can follow the procedure without rediscovering it.

Apply the lifecycleReview sources

Questions to answer

Resolve these before adding tools, frameworks, or automation.

  • What repeated workflow does the Skill make cheaper or more reliable?
  • Which files are instructions, executable helpers, references, or templates?
  • How will an agent know when the Skill applies and when it does not?

Implementation lifecycle

Build the evidence in this order

  1. 01

    Define

    Name one repeatable outcome and document its trigger conditions.

  2. 02

    Build

    Keep the core workflow concise and move optional detail into linked resources.

  3. 03

    Verify

    Run the Skill from a fresh context and validate the produced artifact or action.

Artifacts to maintain

  • SKILL.md
  • scripts/
  • references/
  • assets/

Continue through AgentAtlas

Evidence ledger

Primary documentation behind this guide

Agent Skills

Agent Skills specification

The portable Skill directory structure, required SKILL.md metadata, progressive disclosure, and optional resource folders.

OpenAI

Build skills for ChatGPT and Codex

How Skills are authored, discovered, loaded, and used across OpenAI agent surfaces.

Next topic

Agent architecture