agent skill
In plain wordsA packaged how-to guide for an AI agent — written steps, sometimes with small programs and examples attached — that sits unread until a job of that kind comes up, when the agent opens it.
A reusable bundle of know-how for an AI agentAI agent, n.: An AI system that does not just answer but acts: given a goal, it decides on steps, uses tools such as a browser or code editor, checks the results and keeps going until the job is done., stored as a folder: a SKILL.md file whose YAML front matter gives at least a name and a one-line description and whose body holds Markdown instructions, optionally accompanied by scripts, reference documents and templates. The format is built for progressive disclosureprogressive disclosure, n.: Showing only a short summary first and revealing details when they are actually needed. AI agents use it to avoid filling their limited memory with instructions they may never use., so a skill costs little until it is used: the agent's context permanently holds only the short descriptions, roughly a hundred tokens per skill; the instructions are read when a task seems to match; and supporting files are consulted, or scripts executed, only if the instructions call for them, with just a script's output, not its source, taking space in the context windowcontext window, n.: How much an AI model can read and keep in view at once: the instructions, the conversation so far and any attached documents. Text that does not fit has to be cut, summarized or left out.. One agent can thus carry many specialized competencies (filling PDF forms, applying a company's brand guidelines, running a data pipeline) without crowding its context. Because skills can contain executable code and instructions, untrusted skills are a supply-chain risk and should be audited before installation.
Rather than paste the style guide into every prompt, we packaged it as an agent skill.
Origin
Introduced by Anthropic as 'Agent Skills' on 16 October 2025 for Claude's apps, Claude Code and API (engineering post by Barry Zhang, Keith Lazuka and Mahesh Murag), and published as an open standard on 18 December 2025; OpenAI, Microsoft, Google and dozens of other agent products adopted it. A precursor is the executable skill library of the Minecraft agent Voyager (2023).
Usage note
Skills complement the Model Context ProtocolModel Context Protocol, prop. n.: An open standard that lets AI assistants plug into outside apps and data — files, calendars, databases, business software — through one common connector instead of a custom link for every pairing; often likened to a USB-C port for AI.: MCP connects an agent to external tools and data, while a skill teaches it a procedure, possibly using those tools. Not to be confused with the 'skills' listed in an Agent2Agent protocolAgent2Agent protocol, prop. n.: An open standard that lets AI agents built by different companies find each other, say what they can do and work together on tasks, much as email lets different mail systems exchange messages. Agent Card, which describe an agent's capabilities to other agents.
Related
- progressive disclosureprogressive disclosure, n.: Showing only a short summary first and revealing details when they are actually needed. AI agents use it to avoid filling their limited memory with instructions they may never use.
- agent instruction fileagent instruction file, n.: A plain-text file kept in a software project that tells AI coding agents how the project works — how to build and test it, which conventions to follow, what to avoid — read at the start of every session.
- Model Context ProtocolModel Context Protocol, prop. n.: An open standard that lets AI assistants plug into outside apps and data — files, calendars, databases, business software — through one common connector instead of a custom link for every pairing; often likened to a USB-C port for AI.
- context engineeringcontext engineering, n.: Managing an AI model's reading material: choosing, trimming and arranging the instructions, documents, tool results and memories it is given on every turn, so that it has what the task needs and little else.
- subagentsubagent, n.: A helper AI agent that a main agent starts to handle one piece of a job; it works in its own separate workspace and reports back a short summary when it is done.
- AI supply chain securityAI supply chain security, n.: Protecting everything an AI system is built from — downloaded models, datasets, code libraries and tools — against tampering, so a poisoned or fake part cannot sneak into the finished product.
- fine-tuningfine-tuning, n.: Taking an AI model that has already learned a lot and training it a little more on a smaller, focused set of examples so it gets better at one job or behaves a certain way.
Sources
- Zhang, B., Lazuka, K. & Murag, M. (2025). Equipping agents for the real world with Agent Skills. Anthropic Engineering, 16 October 2025. anthropic.com/engineering/equipping-agents-for-the-real-world-with-agent-skills
- Agent Skills. Overview and specification (open format). agentskills.io
Mentioned in
Model Context ProtocolModel Context Protocol, prop. n.: An open standard that lets AI assistants plug into outside apps and data — files, calendars, databases, business software — through one common connector instead of a custom link for every pairing; often likened to a USB-C port for AI., agent instruction fileagent instruction file, n.: A plain-text file kept in a software project that tells AI coding agents how the project works — how to build and test it, which conventions to follow, what to avoid — read at the start of every session., Claude CodeClaude Code, prop. n.: A coding assistant from Anthropic that works as an agent: given a programming task, it explores the project, changes the code and runs programs to check its work, from the web, a desktop app, an editor or a terminal., OpenClawOpenClaw, prop. n.: A free, open-source AI helper that lives on a computer you control; you send it requests by text message, through apps such as Telegram or Signal, and it acts for you in your email, calendar and files., externalizationexternalization, n.: Moving an AI agent's know-how and memory out of the model itself and into files, tools and rules around it, so the whole system can do more than the model alone., llms.txtllms.txt, n.: A short guide file that a website puts at a standard address to help AI assistants find and understand its most useful pages., progressive disclosureprogressive disclosure, n.: Showing only a short summary first and revealing details when they are actually needed. AI agents use it to avoid filling their limited memory with instructions they may never use.