# Claude Has Skills

> A Claude Skill is a folder with a SKILL.md file: a name, a description of when to use it, and instructions Claude loads only when the task matches. Find official examples in anthropics/skills; write one by naming a single recurring job and describing exactly when to use it.

Last source check: 2026-09-19
Affiliation: Independent publication, not affiliated with Anthropic.

## The Folder
URL: https://claudehasskills.com/folder/
Signature experience: a labeled editorial SKILL.md folder you can open. Click SKILL.md to read name, description, and instructions as a teaching specimen. Not a copied Anthropic skill. Official examples remain in anthropics/skills. Portable format: https://convergentskills.com/skill-folder/. Claude Code practice: https://claudeshipscode.com/.

## Free Claude Skills Tools
URL: https://claudehasskills.com/tools/
Browser-only tools for finding Claude Skill examples, matching workflows to skill patterns, and testing trigger descriptions. No account, backend, signup, or upload endpoint is involved.

### Claude Skills Directory Tool
URL: https://claudehasskills.com/tools/skills-directory/
Description: Search and filter notable public Claude Skill examples from the official Anthropic skills repository, with category notes and source links.
Updated: 2026-09-19
Key fact: 19 - official folders. The directory uses the current `anthropics/skills` GitHub API listing checked on 2026-09-19 during this build.
Key fact: 0 - uploads. Search, filters, and source links run in browser JavaScript. Nothing is sent to a server by this page.
Key fact: Pattern - not endorsement. Use these entries as packaging patterns, then review each skill before installing or adapting it.
How to use:
- Search for the task family or file type you care about.
- Filter by category to compare skills with similar packaging patterns.
- Open the source link and review SKILL.md before installing or copying any pattern.
FAQ:
Q: Is this a complete marketplace of every Claude Skill?
A: No. It is a curated directory of notable public examples, anchored to the official anthropics/skills repository and editorially grouped for study.
Q: Does filtering the directory send my search anywhere?
A: No. The directory is a static page; filtering happens in your browser.
Q: Should I install a skill directly from a directory link?
A: Only after review. Read SKILL.md, inspect scripts and references, and confirm the skill fits your Claude surface and data policy.

### Claude Skill Idea Matcher
URL: https://claudehasskills.com/tools/skill-matcher/
Description: Describe a workflow with toggles and get a browser-only recommendation for existing Claude Skill patterns, adjacent tools, or a new custom skill.
Updated: 2026-09-19
Key fact: Heuristic - pattern matcher. The matcher suggests skill families and next steps. It does not validate a real skill package.
Key fact: 3 - sibling tools. It points to the Convergent Skills generator, description linter, and Skill-vs-MCP decider instead of duplicating them.
Key fact: Local - browser only. Workflow text and toggles stay in the page. No backend, account, or upload is involved.
How to use:
- Describe the workflow in one or two sentences.
- Select the task signals: documents, data, code, browser, media, design, communications, external systems, repeatability, and risk.
- Read the ranked recommendation and follow the links to examples, generator, linter, or MCP decider.
FAQ:
Q: Does the matcher create a SKILL.md file?
A: No. It chooses likely skill families and next steps. Use the linked Convergent Skills scaffold generator when you are ready to draft SKILL.md.
Q: When should the matcher recommend MCP instead of a skill?
A: If the main need is authenticated live system access, stateful operations, or reusable external tools, start with MCP or an existing integration, then add a skill only for procedure.
Q: Can I use the output for Claude.ai and Claude Code?
A: Use the recommendation as a planning aid, then verify packaging and installation rules for the Claude surface you intend to use.

### Claude Skill Trigger Tester
URL: https://claudehasskills.com/tools/trigger-tester/
Description: Paste a SKILL.md description and sample user phrases to estimate which phrases would plausibly trigger the skill.
Updated: 2026-09-19
Key fact: 3 - phrase groups. Test positive, near-miss, and safety prompts before relying on a skill description.
Key fact: Aid - not Claude itself. The tester uses transparent token and signal heuristics; it cannot predict Anthropic runtime behavior exactly.
Key fact: 0 - network calls. Descriptions and phrases are scored locally in browser JavaScript.
How to use:
- Paste the skill description from SKILL.md frontmatter.
- Add one sample user phrase per line, including positive, near-miss, and safety cases.
- Review likely, weak, and unlikely classifications, then tighten the description or eval set.
FAQ:
Q: Is this the same as the Convergent Skills description linter?
A: No. The linter improves the description text. This tester compares that description against sample user phrases to reveal likely trigger gaps.
Q: Can this guarantee how Claude will choose a skill?
A: No. It is a transparent heuristic for authoring and evaluation. Always test in the Claude surface where the skill will run.
Q: What phrases should I test?
A: Use real user wording, adjacent tasks that should not trigger, and risky prompts that should make the skill pause or ask for confirmation.

### Official public Claude Skill examples in the directory tool
- skill-creator (Creator helper): Guides a user through designing, drafting, and testing a reusable skill package. Source: https://github.com/anthropics/skills/tree/main/skills/skill-creator
- docx (Documents and data): Shows how a document-format skill can combine prose guidance with deterministic helpers. Source: https://github.com/anthropics/skills/tree/main/skills/docx
- pdf (Documents and data): Demonstrates PDF-focused workflows where scripts and format constraints matter. Source: https://github.com/anthropics/skills/tree/main/skills/pdf
- pptx (Documents and data): Packages presentation creation and editing guidance around repeatable slide workflows. Source: https://github.com/anthropics/skills/tree/main/skills/pptx
- xlsx (Documents and data): Represents spreadsheet workflows where formulas, sheets, and structured outputs need care. Source: https://github.com/anthropics/skills/tree/main/skills/xlsx
- doc-coauthoring (Communications and collaboration): Shows how a skill can coordinate collaborative document drafting and revision behavior. Source: https://github.com/anthropics/skills/tree/main/skills/doc-coauthoring
- internal-comms (Communications and collaboration): Packages organization-style communication guidance for repeatable internal messages. Source: https://github.com/anthropics/skills/tree/main/skills/internal-comms
- slack-gif-creator (Communications and collaboration): Combines creative output constraints with a specific workplace communication artifact. Source: https://github.com/anthropics/skills/tree/main/skills/slack-gif-creator
- frontend-design (Design and creative): Sets a concrete frontend design standard and anti-pattern list for production UI work. Source: https://github.com/anthropics/skills/tree/main/skills/frontend-design
- brand-guidelines (Design and creative): Illustrates how brand rules, colors, typography, and approval language can be packaged. Source: https://github.com/anthropics/skills/tree/main/skills/brand-guidelines
- canvas-design (Design and creative): Shows a skill aimed at producing polished visual documents and image-like artifacts. Source: https://github.com/anthropics/skills/tree/main/skills/canvas-design
- theme-factory (Design and creative): Demonstrates reusable theme generation as a repeatable creative procedure. Source: https://github.com/anthropics/skills/tree/main/skills/theme-factory
- algorithmic-art (Design and creative): Packages creative coding constraints for generative art workflows. Source: https://github.com/anthropics/skills/tree/main/skills/algorithmic-art
- web-artifacts-builder (Design and creative): Packages interactive web artifact creation patterns and quality bars. Source: https://github.com/anthropics/skills/tree/main/skills/web-artifacts-builder
- claude-api (Developer workflow): Shows how API-specific guidance can be packaged for developers building with Claude. Source: https://github.com/anthropics/skills/tree/main/skills/claude-api
- mcp-builder (Developer workflow): Guides MCP-related engineering work while keeping the protocol/tool boundary visible. Source: https://github.com/anthropics/skills/tree/main/skills/mcp-builder
- webapp-testing (Developer workflow): Packages browser QA behavior, interaction checks, and verification expectations. Source: https://github.com/anthropics/skills/tree/main/skills/webapp-testing
- academy-guide (Conversational moment): Recommends matching Claude Academy courses and tutorials when a user is learning a Claude feature rather than working mid-task. Source: https://github.com/anthropics/skills/tree/main/skills/academy-guide
- discernment-nudge (Conversational moment): Appends two or three checking questions after a substantive answer the user may act on, at most once per conversation. Source: https://github.com/anthropics/skills/tree/main/skills/discernment-nudge

## What Are Claude Skills?
URL: https://claudehasskills.com/what-are-claude-skills/
Description: A Claude Skill is a folder with a SKILL.md file that Claude loads only when your task matches it. Here is how to find, install, and write one.
Updated: 2026-09-19

Quotable fact: 2 - required frontmatter fields. The portable format requires a skill `name` and `description`; everything below the frontmatter is task guidance.
Quotable fact: 3 - main Claude surfaces. Skills work in Claude apps, Claude Code, and the Claude API, but the installation path differs on each surface.

### Short definition
A Claude Skill is a folder containing a SKILL.md file: YAML frontmatter with a name and a description that says when to use it, followed by the instructions Claude should follow. Optional subfolders hold scripts, references, templates, and assets that Claude reads only when the task calls for them.
Claude scans the descriptions of available skills while it works, loads the matching one, and then pulls in deeper files as needed. That is the whole mechanism. A skill is not just a prompt snippet, because it can carry files and executable helpers, and it is not a remote service, because the unit is a package you can open and read.

### How to find, install, or write one
Goal Do this Find one Read the official anthropics/skills repository, or browse the directory tool on this site. Anthropic also publishes a directory of partner-built skills. Install in Claude apps Turn Skills on in Settings. Available on Pro, Max, Team, and Enterprise; Team and Enterprise admins must enable Skills organization-wide first. Install in Claude Code Drop the folder in ~/.claude/skills for personal use or .claude/skills in a repository for the team, or install a plugin from the anthropics/skills marketplace. Use with the API Attach skills to Messages API requests and manage versions through the /v1/skills endpoint. Skills require the code execution tool beta, which supplies the sandbox they run in. Write one Name one recurring job, write a description that says when to use it, then add instructions. The authoring guide walks the full loop.
The fastest way to write a first skill is to ask Claude to do it with the official skill-creator skill, which interviews you about the workflow and generates the folder. Review what it produces before sharing it.

### Why Skills exist
General instructions are brittle when the task has many steps, special file formats, house style rules, or repeated quality checks. Skills let a team put that procedure in one named package, so a future agent run can load the same guidance without forcing the user to paste a long prompt every time.
This is especially valuable for document conversion, brand-guideline work, visual design workflows, internal communications, API usage patterns, and other tasks where "do the job our way" matters as much as the raw model capability.

### Where they run
Claude Code supports skills from three broad locations: project skills in .claude/skills, personal skills in ~/.claude/skills, and plugin-distributed skills from installed marketplaces. Claude apps expose Anthropic-managed skills and custom skills through settings for eligible plans. The Claude API exposes skills as uploadable, versioned packages for Messages API workflows.
That means the same concept spans several product surfaces, but you should still read the current docs for the surface you are using. Claude Code path rules are not the same thing as Claude app upload rules, and API packaging has its own lifecycle.

### Which model runs a skill
The skill format is model-agnostic: the same folder works across the current lineup. As of September 2026 that lineup is Claude Opus 5 as the general default, Claude Fable 5.1 for demanding reasoning and long-horizon agentic runs, Claude Sonnet 5 for the speed and intelligence balance, and Claude Haiku 4.5 as the fastest and cheapest option.
Model choice still changes behavior in two ways worth testing. Stronger models follow long procedures and multi-file routing more faithfully, so a skill that needs Opus 5 or Fable 5.1 to work should say so in its instructions. Reasoning effort also differs by surface — Fable 5.1 defaults to high effort in Claude Code and medium in the Claude apps — which can change how much of a bundled reference the model actually reads. For model selection and pricing, see Claude Central ; this guide stays on packaging.

### What a Skill is not
A skill is not an MCP server. MCP connects Claude to external tools and data sources through a protocol. A skill can explain how to use a tool, or include a script, but it does not turn into a live connector just because it exists.
A skill is not a complete product integration by itself. It can encode a workflow, provide templates, and run bundled helpers, but credentials, permissions, network access, and external systems still need ordinary engineering review.

### FAQ
Q: Is this site affiliated with Anthropic?
A: No. Claude Has Skills is an independent editorial guide. It links to Anthropic and Agent Skills primary sources so readers can verify product details directly.
Q: Does a skill replace normal prompting?
A: No. A skill packages repeatable instructions and resources so the agent can load them when a task matches. The user still asks for a task, and the agent still decides what to read and run.
Q: When should I create a skill?
A: Create one when the same specialized workflow is likely to repeat and the agent needs durable instructions, examples, scripts, or reference material to perform it consistently.
Q: What is the minimum skill?
A: A folder containing a `SKILL.md` file with valid YAML frontmatter, including `name` and `description`, followed by clear instructions.
Q: Where do I find Claude Skills?
A: Start with the official `anthropics/skills` repository on GitHub, which holds Anthropic example skills, and the <a href="/tools/skills-directory/">directory tool</a> on this site. Anthropic also publishes a directory of partner-built skills for Claude apps.
Q: Which Claude model do Skills work with?
A: All of the current lineup. The format is the same for Claude Opus 5, Fable 5.1, Sonnet 5, and Haiku 4.5, though longer procedures are followed more reliably by Opus 5 and Fable 5.1.
Q: Do Skills cost extra?
A: There is no separate skill fee. You pay ordinary model usage, plus a loaded skill adds tokens to the request and API skills require the code execution tool.

### Primary sources
- What are Skills? (Claude Help Center): https://support.claude.com/en/articles/12512176-what-are-skills
- Introducing Agent Skills (Anthropic News): https://www.anthropic.com/news/skills
- Claude Code Skills docs (Anthropic Docs): https://code.claude.com/docs/en/skills
- Skills API guide (Anthropic Docs): https://platform.claude.com/docs/en/build-with-claude/skills-guide
- Agent Skills specification (Agent Skills): https://agentskills.io/specification
- Choosing a Claude model (Anthropic Docs): https://platform.claude.com/docs/en/about-claude/models/choosing-a-model

## Claude Skills Directory
URL: https://claudehasskills.com/directory/
Description: A curated directory of official Anthropic example skills, what each category teaches, and how to evaluate community skills.
Updated: 2026-09-19

Quotable fact: 19 - official example folders reviewed. The Anthropic repository listing reviewed on 2026-09-19 exposed nineteen top-level skill folders under `skills/`.
Quotable fact: 0 - unverified marketplace claims. This directory lists official examples and evaluation criteria; it does not present scraped community packages as endorsed.

### How to use this directory
Use the official repository as a pattern library. The important question is not "Which skill should I copy?" but "Which packaging pattern resembles my task?" A document-processing skill teaches different lessons than a design skill, a web-testing skill, or an API helper skill.
For each skill you inspect, look at the trigger description, the amount of instruction in SKILL.md, the use of references, whether scripts are deterministic, and how much human judgment the skill expects the model to apply.

### Official examples worth studying
Example What it teaches Review angle skill-creator How to guide a user through creating a reusable skill. Study trigger wording and creator workflow. docx, pdf, pptx, xlsx How complex document skills combine instructions, scripts, and format-specific helpers. Review how much logic lives in scripts instead of prose. frontend-design How an aesthetic standard can steer an agent without becoming a generic prompt. Look for concrete anti-patterns and decision rules. webapp-testing How to package browser QA procedures. Check tool assumptions and reproducibility. mcp-builder How a skill can guide an engineering workflow adjacent to external tools. Keep the boundary clear between guidance and the MCP server itself. discernment-nudge, academy-guide How a skill can fire on a conversational moment rather than a file type, and cap itself to once per conversation. Read the description as a routing rule; these are the clearest examples of exclusion wording.
The repository listing also includes creative and enterprise examples such as algorithmic-art, brand-guidelines, canvas-design, claude-api, doc-coauthoring, internal-comms, slack-gif-creator, theme-factory, and web-artifacts-builder. These are useful because they show that Skills are not only for coding; they are a general packaging mechanism for repeatable agent behavior.
New since the July review: academy-guide, which recommends matching Claude Academy courses when a user is learning rather than mid-task, and discernment-nudge, which appends a few checking questions after a substantive answer. Both are worth reading as description craft. Their frontmatter spends most of its length on when not to fire, which is exactly the discipline broad skills usually lack.

### Skills you do not install
Not every skill in circulation is a folder you can download. Anthropic ships managed skills inside the Claude apps for document creation, and the educator offering announced in July 2026 includes teaching skills for verified US K-12 teachers that arrive with the account rather than through a repository. If you are looking for that side of Claude, Claude Helps covers the everyday and classroom surfaces; this directory stays on packages you can inspect.
The same applies to partner skills for tools like Box, Canva, and Notion. They are distributed through the Claude apps, so the reviewable artifact is the vendor integration and its permissions, not a SKILL.md you can read line by line.

### Rules for community skills
Treat a community skill like a small software dependency. Read SKILL.md, inspect every script, check whether it reaches the network, and decide whether the instructions can exfiltrate private data through ordinary agent behavior. A skill with no scripts can still be risky if it tells the agent to paste sensitive material into a third-party service.
Prefer skills with narrow names, direct descriptions, visible examples, changelogs, and test prompts. Be cautious with broad skills that promise to "make Claude better at everything"; they are harder to trigger predictably and harder to audit.

### What belongs elsewhere
General Claude Code best practices belong on a Claude Code engineering site. Context windows, memory, and compaction belong in context and memory guides. This directory stays focused on skill packages: what exists, what pattern it demonstrates, and what to check before using it.
If a package mainly exposes external tools, it may be an MCP server or plugin rather than a skill. If it mainly stores project facts, it may belong in project instructions or memory. Use the boundary guide before adding it to your own directory.

### FAQ
Q: Is this site affiliated with Anthropic?
A: No. Claude Has Skills is an independent editorial guide. It links to Anthropic and Agent Skills primary sources so readers can verify product details directly.
Q: Does a skill replace normal prompting?
A: No. A skill packages repeatable instructions and resources so the agent can load them when a task matches. The user still asks for a task, and the agent still decides what to read and run.
Q: Are the official example skills endorsements for production use?
A: No. The repository itself frames the skills as demonstrations and references. Production use still needs local testing and security review.
Q: Should I install every useful-looking skill?
A: No. Install only skills that match recurring work. Too many broad skills make triggering and review harder.
Q: How many official Anthropic example skills are there?
A: The `anthropics/skills` repository exposed nineteen top-level folders under `skills/` when this page was last checked on September 19, 2026. That count changes, so verify against the repository listing before citing it.

### Primary sources
- anthropics/skills repository (GitHub): https://github.com/anthropics/skills
- anthropics/skills current directory listing (GitHub API): https://api.github.com/repos/anthropics/skills/contents/skills
- Introducing Agent Skills (Anthropic News): https://www.anthropic.com/news/skills
- Using Skills in Claude (Claude Help Center): https://support.claude.com/en/articles/12512180-use-skills-in-claude
- Claude for Teachers (Anthropic News): https://www.anthropic.com/news/claude-for-teachers
- Best practices for skill creators (Agent Skills): https://agentskills.io/skill-creation/best-practices

## Skill Anatomy: SKILL.md, References, Scripts, Assets
URL: https://claudehasskills.com/skill-anatomy/
Description: A portable Claude Skill folder anatomy guide covering required frontmatter, instruction design, references, scripts, and assets.
Updated: 2026-09-19

Quotable fact: 1 - required front door. A portable skill is discovered through its `SKILL.md`; optional files should support that front door rather than replace it.
Quotable fact: 4 - common layers. `SKILL.md`, `references/`, `scripts/`, and `assets/` form a practical four-layer skill structure.

### The minimal folder
The minimal skill is a directory containing SKILL.md. The frontmatter names the skill and describes when it should be used. The Markdown body gives the agent the operating procedure: steps, checks, examples, constraints, and references to any bundled files.
my-skill/ SKILL.md
That small shape is enough for many editorial or review workflows. Add more files only when they make the skill easier to maintain, safer to execute, or more reproducible.

### Frontmatter: name and description
The name should be stable, lowercase, and easy to refer to. The description is not marketing copy; it is a trigger contract. It should tell the agent and user exactly which task calls for the skill and which nearby tasks do not.
A weak description says "helps with design." A stronger one says "Use when creating production-grade frontend interfaces that need distinctive visual design and avoidance of generic AI aesthetics." The second version gives the agent a real matching boundary.

### References: deep material on demand
Put long policy, domain, or API notes in references/ instead of stuffing everything into SKILL.md. The skill entrypoint should route the agent to the right reference when needed. This keeps the first loaded context small and lets the agent read deeper only after it knows the task variant.
Good reference files are named for decisions: brand-voice.md, api-auth.md, review-rubric.md, or invoice-rules.md. Avoid vague names such as notes.md when the folder may grow.

### Scripts and assets
Use scripts/ when the workflow has deterministic work that should not be retyped from memory: rendering PDFs, validating schemas, extracting metadata, checking a fixture, or running a repeatable conversion. A script should be narrow, inspectable, and documented by the skill instructions.
Use assets/ for templates, example files, visual materials, boilerplate, and other reusable inputs. Assets are useful when the output must match a format that prose alone cannot describe reliably.

### Visual model
A practical skill folder keeps SKILL.md as the front door and moves heavier materials into named subfolders.

### FAQ
Q: Is this site affiliated with Anthropic?
A: No. Claude Has Skills is an independent editorial guide. It links to Anthropic and Agent Skills primary sources so readers can verify product details directly.
Q: Does a skill replace normal prompting?
A: No. A skill packages repeatable instructions and resources so the agent can load them when a task matches. The user still asks for a task, and the agent still decides what to read and run.
Q: Do I need scripts in every skill?
A: No. Add scripts only when repeatable execution is safer or more accurate than prose instructions.
Q: Should `SKILL.md` contain every detail?
A: No. Keep `SKILL.md` focused on activation and routing. Move long variant-specific material into references.

### Primary sources
- Agent Skills specification (Agent Skills): https://agentskills.io/specification
- Claude Code Skills docs (Anthropic Docs): https://code.claude.com/docs/en/skills
- Best practices for skill creators (Agent Skills): https://agentskills.io/skill-creation/best-practices
- anthropics/skills repository (GitHub): https://github.com/anthropics/skills

## How to Create a Claude Skill
URL: https://claudehasskills.com/authoring-guide/
Description: A practical, source-backed authoring workflow for creating a focused Claude Skill that triggers reliably and stays maintainable.
Updated: 2026-09-19

Quotable fact: 5 - authoring passes. Scope, trigger, instruct, package, and evaluate before treating a skill as reusable.
Quotable fact: 1 - primary job. A strong skill does one recurring job well; broad omnibus skills are harder to trigger and audit.

### 1. Choose one recurring job
Start with the work, not the format. A good skill candidate is a task you can name in one sentence, repeat across projects, and judge with a recognizable definition of done. "Create investor-ready board deck charts from this spreadsheet" is a candidate. "Help with business" is not.
Write the non-goals beside the goal. If the skill writes launch copy, say whether it should also generate images, post to social media, or inspect analytics. Clear exclusions improve both safety and trigger accuracy.

### 2. Write the trigger description
The description field does double duty: it tells humans what the skill is for and gives the agent a matching rule. Include the task, the domain, and the signal that should cause activation. Leave out grand claims, adjectives, and implementation details that do not help matching.
--- name: invoice-audit description: Use when reviewing vendor invoices against contract terms, purchase orders, and known exception rules. ---
If you cannot write a specific description, the skill is probably not scoped tightly enough.

### 3. Write the procedure
Use imperative steps. Tell the agent what to inspect first, which files to open, which assumptions to avoid, and what output shape to produce. Give examples only when they remove ambiguity. For complicated tasks, route to a reference file instead of overloading the entrypoint.
Strong procedures include stop conditions. A skill that might touch credentials, legal claims, medical advice, external deploys, or customer data should tell the agent when to ask for confirmation or perform a read-only review.

### 4. Package references, scripts, and assets
Add files because they reduce error. If the agent needs a current schema, put the schema in the skill. If the job needs a parser, put a script in scripts/. If the output must match a format, include a template in assets/.
Every bundled file should have a reason to exist. If a reference file duplicates the same instruction in three places, cut it down. If a script makes a network call or writes to disk, document that behavior in SKILL.md so reviewers can spot it quickly.

### 5. Evaluate before sharing
Test positive prompts, near-miss prompts, and adversarial prompts. A positive prompt should trigger the skill. A near-miss prompt should not. An adversarial prompt should not make the skill ignore its own safety boundaries.
Keep a small eval log next to the skill while it is young. The log does not need a full framework on day one; a table of prompt, expected behavior, actual behavior, and fix is enough to improve the description and instructions.

### FAQ
Q: Is this site affiliated with Anthropic?
A: No. Claude Has Skills is an independent editorial guide. It links to Anthropic and Agent Skills primary sources so readers can verify product details directly.
Q: Does a skill replace normal prompting?
A: No. A skill packages repeatable instructions and resources so the agent can load them when a task matches. The user still asks for a task, and the agent still decides what to read and run.
Q: How long should `SKILL.md` be?
A: Long enough to route and execute the core procedure, but short enough that variant-specific details can live in named references.
Q: Can I create a skill from an existing prompt?
A: Yes, but rewrite it as an operating procedure. Add trigger metadata, remove one-off context, and package any reusable files.

### Primary sources
- Creating custom Skills (Claude Help Center): https://support.claude.com/en/articles/12512198-how-to-create-custom-skills
- Agent Skills quickstart (Agent Skills): https://agentskills.io/skill-creation/quickstart
- Best practices for skill creators (Agent Skills): https://agentskills.io/skill-creation/best-practices
- Agent Skills specification (Agent Skills): https://agentskills.io/specification
- anthropics/skills repository (GitHub): https://github.com/anthropics/skills

## Claude Code Skills: Where They Install
URL: https://claudehasskills.com/claude-code/
Description: The three places a Claude Code skill can live, how project and personal scope differ, and which path to pick.
Updated: 2026-09-19

Quotable fact: 3 - skill sources in Claude Code. Project, personal, and plugin-distributed skills serve different scopes in Claude Code.
Quotable fact: .claude/skills - project skill path. Project skills belong with the repository so collaborators can review and version them.

### Three scopes
Scope Location Use when Project .claude/skills in the repository The workflow is specific to this codebase and the team should review it. Personal ~/.claude/skills You want the skill in every project without committing it anywhere. Plugin Installed marketplace, including anthropics/skills You want maintained skills you did not write, updated by their publisher.
Use project skills when the workflow belongs to a repository. They should live under .claude/skills so the team can review changes, version them, and keep task behavior close to the code or documentation it affects.
Use personal skills for private workflows you want across projects. These belong in your user-level Claude configuration, not in a shared repository. Use plugin-distributed skills when the skill comes from an installed Claude Code plugin or marketplace source; the announcement points at the anthropics/skills marketplace as the starting point.

### When to create a project skill
Create a project skill when the repository has a repeated local practice: release checks, architecture review, design QA, data migration rehearsal, support-log triage, or documentation conversion. The skill should encode the project-specific rules that generic Claude Code guidance cannot know.
Do not use a project skill as a dumping ground for every team preference. Stable repository facts may belong in project instructions; specialized procedures with supporting files belong in skills.

### Invocation and triggering
Users can mention a skill by name, but the stronger design is to make the description clear enough that the agent can select it when the task matches. The description should say when to use the skill, not just what the skill is called.
If a skill fails to trigger, revise the description before adding paragraphs of instructions. Trigger matching depends on the skill being described in the same language users naturally use for the task.

### Version control and review
Treat project skills like code-adjacent configuration. Review diffs, avoid hidden network behavior, keep scripts executable and small, and document dependencies. A skill that changes deploy behavior deserves the same scrutiny as a shell script in the repo.
This page stops at installation and scope on purpose. Using skills as part of a Claude Code engineering practice — how they sit alongside hooks, subagents, CI runs, and MCP servers — belongs to Claude Ships Code , and the Agent SDK offers the same skill support for custom agents.

### FAQ
Q: Is this site affiliated with Anthropic?
A: No. Claude Has Skills is an independent editorial guide. It links to Anthropic and Agent Skills primary sources so readers can verify product details directly.
Q: Does a skill replace normal prompting?
A: No. A skill packages repeatable instructions and resources so the agent can load them when a task matches. The user still asks for a task, and the agent still decides what to read and run.
Q: Should I commit project skills?
A: Usually yes, if they define shared repository workflows. Committing them makes review and team alignment possible.
Q: Can a Claude Code plugin provide skills?
A: Yes. Claude Code can receive skills from installed plugin sources, including the `anthropics/skills` marketplace, which should still be reviewed like any other dependency.
Q: Where can I read about using skills in a Claude Code workflow?
A: This site covers the skill package itself. For Claude Code engineering practice around skills, hooks, agents, and CI, see <a href="https://claudeshipscode.com/">Claude Ships Code</a>.

### Primary sources
- Claude Code Skills docs (Anthropic Docs): https://code.claude.com/docs/en/skills
- Introducing Agent Skills (Anthropic News): https://www.anthropic.com/news/skills
- anthropics/skills repository (GitHub): https://github.com/anthropics/skills
- Best practices for skill creators (Agent Skills): https://agentskills.io/skill-creation/best-practices

## Claude Apps and API Skills
URL: https://claudehasskills.com/claude-ai-api/
Description: What changes when Claude Skills are used in the Claude apps or the Claude API instead of local Claude Code.
Updated: 2026-09-19

Quotable fact: /v1/skills - API endpoint. API skills are versioned package uploads attached to Messages API requests, not local folder discovery.
Quotable fact: Pro and up - app availability. Skills are available to Pro, Max, Team, and Enterprise users; Team and Enterprise admins enable them organization-wide first.

### Why the surface matters
The same skill idea travels across Claude products, but installation is not universal. A local folder path matters in Claude Code. A UI upload matters in Claude.ai. An API upload and identifier matter in a Messages API workflow.
When documenting a skill for users, say which surface you tested. "Works as a Claude Code project skill" is more useful than a vague "works with Claude" claim.

### Claude apps
In the Claude apps, skills are aimed at end-user workflows such as files, documents, design artifacts, and custom organizational procedures. Enable them in Settings; they are available on Pro, Max, Team, and Enterprise, and Team or Enterprise admins must switch Skills on for the organization before individual users see them. The Help Center is the right source for upload instructions and product limitations because those rules change faster than the portable file format.
Activation is automatic rather than manual. Claude picks the matching skill from the task and shows it in its chain of thought, which is a useful debugging signal when a skill fires at the wrong time. If the workflow needs local repo access, shell commands, or codebase-specific automation, Claude Code may be the better surface.

### Claude API
In the API, a skill is part of an application workflow. Skills can be attached to Messages API requests, and the /v1/skills endpoint gives programmatic control over custom skill versioning and management; the Claude Console covers the same create, view, and upgrade steps interactively. Skills on this surface require the code execution tool beta, which provides the sandbox they run in — so a skill is not a drop-in addition to an existing request if that tool is not enabled.
Pick the model for the job rather than assuming the skill compensates. Opus 5 is the general default, Fable 5.1 suits long-horizon agentic runs that load several bundled references, and Sonnet 5 is the balance choice for high-volume skill calls. API usage also raises concerns the skill text cannot solve: package versioning, tenant isolation, file retention, audit logs, and explicit testing around tool calls or generated outputs.

### Portability expectations
Portable skill design still helps even when product-specific packaging differs. A clear SKILL.md, named references, deterministic scripts, and small assets are easier to adapt across surfaces than a giant prompt pasted into one product.
If a skill depends on a Claude Code-only path, a local executable, or a private service, document that dependency in the first screen of the skill so API and Claude.ai users do not assume it is portable.

### FAQ
Q: Is this site affiliated with Anthropic?
A: No. Claude Has Skills is an independent editorial guide. It links to Anthropic and Agent Skills primary sources so readers can verify product details directly.
Q: Does a skill replace normal prompting?
A: No. A skill packages repeatable instructions and resources so the agent can load them when a task matches. The user still asks for a task, and the agent still decides what to read and run.
Q: Can a skill work in every Claude product?
A: Conceptually yes, but practically only if its dependencies and packaging match the product surface. Always verify the target surface.
Q: Where should API-specific behavior live?
A: Keep API lifecycle, authentication, and tenant behavior in application code and docs. Let the skill carry the task procedure and resources.
Q: Which plans include Skills in the Claude apps?
A: Pro, Max, Team, and Enterprise. On Team and Enterprise an admin must enable Skills organization-wide before users can turn them on in Settings.
Q: Do API skills need any other beta feature?
A: Yes. Skills in the API require the code execution tool beta, which supplies the sandbox the skill runs in.

### Primary sources
- Skills API guide (Anthropic Docs): https://platform.claude.com/docs/en/build-with-claude/skills-guide
- Introducing Agent Skills (Anthropic News): https://www.anthropic.com/news/skills
- Using Skills in Claude (Claude Help Center): https://support.claude.com/en/articles/12512180-use-skills-in-claude
- Creating custom Skills (Claude Help Center): https://support.claude.com/en/articles/12512198-how-to-create-custom-skills
- Agent Skills specification (Agent Skills): https://agentskills.io/specification

## Claude Skills vs MCP, Projects, Plugins, and Instructions
URL: https://claudehasskills.com/skills-vs-mcp-projects-instructions/
Description: A boundary guide for deciding whether a workflow belongs in a Skill, MCP server, project instructions, plugin, or ordinary prompt.
Updated: 2026-09-19

Quotable fact: 1 - routing question. Ask what must be packaged: procedure, external capability, persistent facts, or one-time context.
Quotable fact: MCP - external connector. MCP is the better fit when the main need is live access to external tools or data.

### Decision table
Need Best fit Why Repeatable procedure with examples, references, or scripts Skill The agent needs packaged know-how. Live external system access MCP or tool The agent needs a connector or function call. Repository-wide facts and preferences Project instructions The agent needs persistent local context. Distribution of commands, hooks, and skills Plugin The package needs multiple capabilities. One-off task context Prompt The information is not reusable yet.

### Skills vs MCP
Use a skill when the valuable thing is procedure: how to review, write, transform, test, or evaluate. Use MCP when the valuable thing is live capability: querying a system, taking an action, retrieving data, or exposing a tool surface from another process.
They can work together. A skill can tell the agent how to use a particular MCP server safely. But the skill is the guide; the MCP server is the connector.

### Skills vs project instructions
Project instructions are always-on background guidance for a repository or workspace. Skills are conditional packages for specialized tasks. If the agent should remember a rule on every turn, it probably belongs in project instructions. If the agent should activate a procedure only for a matching job, it probably belongs in a skill.
A common mistake is placing every team preference into a skill. That makes the skill noisy and weakens its trigger. Put stable project facts where the agent can always see them; put specialized methods in skills.

### Skills vs plugins
A plugin can distribute skills, commands, hooks, or integrations. A skill is one kind of capability that may be packaged by a plugin. When a user says "install this plugin," they may be installing a bundle that includes several skills.
For directory purposes, list the skill by the job it performs, not merely by the plugin that ships it. That makes review easier and prevents plugin branding from hiding the actual behavior.

### FAQ
Q: Is this site affiliated with Anthropic?
A: No. Claude Has Skills is an independent editorial guide. It links to Anthropic and Agent Skills primary sources so readers can verify product details directly.
Q: Does a skill replace normal prompting?
A: No. A skill packages repeatable instructions and resources so the agent can load them when a task matches. The user still asks for a task, and the agent still decides what to read and run.
Q: Can a skill call tools?
A: A skill can instruct the agent to use available tools or scripts, but the tool capability itself is provided by the environment, API, or connector.
Q: Should company policy live in a skill?
A: Only if it supports a specialized repeatable workflow. Always-on policy belongs in a more global instruction surface.

### Primary sources
- Claude Code Skills docs (Anthropic Docs): https://code.claude.com/docs/en/skills
- Tool use overview (Anthropic Docs): https://platform.claude.com/docs/en/agents-and-tools/tool-use/overview
- Model Context Protocol connector docs (Anthropic Docs): https://platform.claude.com/docs/en/agents-and-tools/mcp-connector
- Agent Skills specification (Agent Skills): https://agentskills.io/specification
- Best practices for skill creators (Agent Skills): https://agentskills.io/skill-creation/best-practices

## Claude Skill Examples and Patterns
URL: https://claudehasskills.com/examples/
Description: A pattern catalog for reading official Claude Skill examples and designing your own without copying blindly.
Updated: 2026-09-19

Quotable fact: 6 - pattern families. Document, design, developer, enterprise, creator-helper, and conversational-moment skills demonstrate different packaging choices.
Quotable fact: 0 - blind copies recommended. Examples should be mined for patterns, not copied into production without local review.

### Document and file-format skills
The docx, pdf, pptx, and xlsx examples are the right place to study complex skills. They show how a skill can combine prose instructions with helper scripts and format-specific knowledge. The lesson is not that every skill needs a script; the lesson is that scripts are useful when file formats need deterministic handling.
If your workflow manipulates structured files, start by deciding which operations must be exact. Put exact operations into scripts or templates, and let the model handle judgment, prioritization, and narrative output.

### Design and creative skills
Skills such as frontend-design, brand-guidelines, canvas-design, theme-factory, and algorithmic-art demonstrate taste-setting. They work by naming the aesthetic standard, describing anti-patterns, and giving the agent a vocabulary for decisions.
The risk with creative skills is generic taste language. Replace vague words like "polished" with concrete rules: typography constraints, layout expectations, unacceptable visual tropes, or domain-specific examples.

### Developer workflow skills
Developer skills such as webapp-testing, mcp-builder, and claude-api show how a skill can guide an agent through technical work without becoming a full tutorial. The best versions define when to inspect docs, when to run tools, and what output evidence counts as done.
Keep developer skills close to verifiable commands. A skill that says "test carefully" is weak. A skill that names the exact browser checks, command sequence, expected artifact, and failure-report format is much stronger.

### Enterprise workflow skills
Skills for internal communications, coauthoring, or brand guidelines are useful because they package institutional preferences. The skill can tell the agent how the organization writes, which templates to use, what approvals matter, and how to avoid claims that the organization would not make.
These are also the skills most likely to contain sensitive information. Separate public examples from private policies, and run a security review before sharing a skill outside the team.

### Creator-helper skills
skill-creator is a meta-skill: it helps create skills. Study it to see how an agent can interview a user, translate workflow details into frontmatter, and produce a starter folder. Meta-skills are powerful, but their output still needs review by someone who understands the target task.
A creator-helper skill should produce small, inspectable artifacts. Avoid generators that create huge skill bundles before the workflow has been tested on real prompts.

### Conversational-moment skills
The newest official examples trigger on a moment in the conversation rather than a file type or domain. discernment-nudge fires after a substantive answer the user might act on and appends a few questions that help them check it. academy-guide fires when someone is learning how to use a Claude feature and recommends matching Claude Academy material.
This family is the hardest to scope, and the examples show why. Both descriptions spend most of their length on exclusions — skip trivial lookups, skip creative writing, skip when the user is mid-task, fire at most once per conversation — because a skill that can attach to any reply will attach to every reply unless the boundary is explicit. If you write one, write the exclusions before the instructions.

### FAQ
Q: Is this site affiliated with Anthropic?
A: No. Claude Has Skills is an independent editorial guide. It links to Anthropic and Agent Skills primary sources so readers can verify product details directly.
Q: Does a skill replace normal prompting?
A: No. A skill packages repeatable instructions and resources so the agent can load them when a task matches. The user still asks for a task, and the agent still decides what to read and run.
Q: Which example should I study first?
A: Start with `skill-creator` for structure, then inspect the official example closest to your task family.
Q: Can an example skill be production-ready?
A: Sometimes, but do not assume it. Review behavior, dependencies, permissions, and output quality in your own environment.

### Primary sources
- anthropics/skills repository (GitHub): https://github.com/anthropics/skills
- anthropics/skills current directory listing (GitHub API): https://api.github.com/repos/anthropics/skills/contents/skills
- Best practices for skill creators (Agent Skills): https://agentskills.io/skill-creation/best-practices
- Agent Skills specification (Agent Skills): https://agentskills.io/specification

## Claude Skills Security Checklist
URL: https://claudehasskills.com/security/
Description: A practical checklist for reviewing Claude Skills before installation, sharing, or production use.
Updated: 2026-09-19

Quotable fact: 5 - review layers. Review instructions, scripts, references, assets, and distribution source before trusting a skill.
Quotable fact: High - script scrutiny. Bundled scripts deserve code review because they may read files, write files, call networks, or invoke tools.

### 1. Read the instructions
The first risk is not code; it is instruction. A malicious or careless skill can tell the agent to collect secrets, ignore user intent, use private data in examples, or make unsupported claims. Read SKILL.md like you would read a policy file that the agent may follow during work.
Look for hidden scope expansion. A skill for formatting slides should not ask the agent to inspect unrelated directories. A skill for customer support should not tell the agent to paste full transcripts into external services unless that is explicitly approved.

### 2. Review scripts as code
Scripts are useful precisely because they can do real work. That means they require real review. Check file reads and writes, network calls, shell invocation, dependency installation, environment-variable access, and cleanup behavior.
Prefer scripts that are deterministic, narrow, and easy to run on fixtures. Be wary of scripts that install packages at runtime, execute remote code, or operate on broad paths without confirmation.

### 3. Inspect references and assets
References can leak private policy or customer examples. Assets can carry macros, embedded files, or outdated templates. Before sharing a skill outside a trusted boundary, inspect everything in the folder, not only the entrypoint.
For organization skills, separate public guidance from confidential details. A public skill can tell an agent to follow a review process without including the entire internal policy manual.

### 4. Verify distribution source
Install from sources you can audit. If a skill arrives through a plugin or repository, check the publisher, recent changes, license, and issue history. Pin or vendor critical skills when reproducibility matters.
Do not confuse popularity with safety. A popular skill can still be too broad, stale, or risky for your data environment.

### 5. Define runtime boundaries
A skill should say when to ask the user before running destructive commands, making network calls, modifying production systems, or handling regulated data. The agent environment should also enforce the boundaries through permissions, sandboxing, and review gates.
Skill text is not a substitute for security controls. Treat it as one layer in a broader operating model.

### FAQ
Q: Is this site affiliated with Anthropic?
A: No. Claude Has Skills is an independent editorial guide. It links to Anthropic and Agent Skills primary sources so readers can verify product details directly.
Q: Does a skill replace normal prompting?
A: No. A skill packages repeatable instructions and resources so the agent can load them when a task matches. The user still asks for a task, and the agent still decides what to read and run.
Q: Can a no-script skill be risky?
A: Yes. Instructions alone can cause unsafe behavior if they ask the agent to misuse data, ignore policy, or overreach.
Q: Should third-party skills be allowed in company repos?
A: Only after review. Treat them like dependencies and require an owner, source, version, and rollback path.

### Primary sources
- Agent Skills specification (Agent Skills): https://agentskills.io/specification
- Claude Code Skills docs (Anthropic Docs): https://code.claude.com/docs/en/skills
- anthropics/skills repository (GitHub): https://github.com/anthropics/skills
- Best practices for skill creators (Agent Skills): https://agentskills.io/skill-creation/best-practices

## Evaluate Claude Skills: Triggering, Output, Safety
URL: https://claudehasskills.com/evaluation/
Description: How to test whether a Claude Skill triggers at the right time, produces useful output, and respects safety boundaries.
Updated: 2026-09-19

Quotable fact: 3 - prompt sets. Use positive, near-miss, and adversarial prompts to evaluate trigger behavior.
Quotable fact: 4 - minimum columns. Prompt, expected behavior, actual behavior, and fix are enough for a practical early eval log.

### Trigger evals
A skill that never triggers is dead. A skill that triggers everywhere is dangerous. Build a small prompt set with three groups: tasks that should activate the skill, tasks that sound adjacent but should not, and tasks that try to lure the skill outside its stated boundaries.
Write prompts in the language real users use. If users say "make this match our board deck," the description should probably mention board decks, not only "executive presentation system."

### Output evals
Triggering is only the first test. The output should meet the stated definition of done. If the skill creates a document, inspect format and content. If it reviews code, inspect finding quality and file references. If it runs a script, inspect command output and failure behavior.
Use fixtures whenever possible. A known input with a known desired output gives you a stable regression test when you change the description or procedure.

### Safety evals
Safety evals should test predictable pressure points: secrets, destructive commands, private data, regulated claims, and external services. A finance-analysis skill should not invent compliance claims. A deploy skill should not push production changes without confirmation.
The goal is not to make the skill paranoid. The goal is to make the stop conditions explicit enough that the agent can preserve momentum on ordinary work and pause on risky work.

### Keep a tiny eval log
Prompt Expected Actual Fix Review this vendor invoice against our contract. Triggers invoice-audit. Triggered. None. Write a friendly invoice reminder. Does not trigger. Triggered. Narrow description to audit/reconciliation. Audit this invoice and email the vendor. Ask before external send. Asked. None.
Keep the log close to the skill while iterating. Once the skill is stable, promote the most important prompts to a formal test harness or review checklist.

### FAQ
Q: Is this site affiliated with Anthropic?
A: No. Claude Has Skills is an independent editorial guide. It links to Anthropic and Agent Skills primary sources so readers can verify product details directly.
Q: Does a skill replace normal prompting?
A: No. A skill packages repeatable instructions and resources so the agent can load them when a task matches. The user still asks for a task, and the agent still decides what to read and run.
Q: How many eval prompts do I need?
A: Start with at least three positive prompts, three near-miss prompts, and three safety prompts. Expand when the skill handles higher-risk work.
Q: What should I change first when a skill misfires?
A: Revise the description first. If the skill triggers correctly but performs poorly, revise the procedure or supporting files.

### Primary sources
- Best practices for skill creators (Agent Skills): https://agentskills.io/skill-creation/best-practices
- Agent Skills quickstart (Agent Skills): https://agentskills.io/skill-creation/quickstart
- Agent Skills specification (Agent Skills): https://agentskills.io/specification
- anthropics/skills repository (GitHub): https://github.com/anthropics/skills

## How Claude Skills Trigger
URL: https://claudehasskills.com/skill-triggering/
Description: A practical guide to writing and testing Claude Skill descriptions so the right user phrases activate the right skill.
Updated: 2026-09-19

Quotable fact: 1 - description contract. The `description` field should say when to use the skill in language close to real user requests.
Quotable fact: 3 - eval groups. Positive, near-miss, and safety prompts expose most early triggering problems.

### Triggering is routing, not magic
Claude Skill triggering starts with the metadata a client can see before loading the full skill. The name identifies the package, but the description carries the routing contract: which user tasks should cause the agent to load the skill and which nearby tasks should not.
This is why description writing belongs in the authoring loop, not at the end. A skill with strong procedures and weak discovery metadata may never be used. A skill with broad discovery metadata may appear in tasks where it adds risk or noise.

### Write in user language
Good descriptions reuse the words users actually type: review invoices, convert PDFs, create board deck charts, test browser flows, draft internal announcements. Bad descriptions rely on internal jargon, broad value claims, or abstract verbs that could match anything.
Prefer "Use when reviewing vendor invoices against contracts and purchase orders" over "Use for finance excellence." The first description contains the task, artifact, and boundary. The second description is impossible to test.

### Evaluate the trigger
Create three prompt groups. Positive prompts should activate the skill. Near-miss prompts should not. Safety prompts should either not activate or activate with an explicit pause, depending on the skill boundary.
For quick iteration, paste the description and sample phrases into the Claude Skill Trigger Tester . Then use the Convergent Skills Description Linter to repair vague or overbroad wording.

### Common triggering failures
Failure Symptom Fix Too broad The skill appears for adjacent work. Add domain objects, artifacts, and exclusions. Too internal Real user wording does not match. Rewrite around phrases users type. Too capability-focused The description says what the skill contains, not when to use it. Start with "Use when..." and name the task. No stop condition Risky prompts look like ordinary work. Add safety boundaries to the procedure and eval prompts.

### A small repair loop
When a skill misfires, revise in this order: description first, examples second, procedure third. If the skill does not trigger, the description is the likely cause. If it triggers but performs badly, the procedure or supporting files are the likely cause.
Keep a tiny eval log beside the skill while it is young. A four-column table with prompt, expected behavior, actual behavior, and fix is enough to make trigger behavior visible before the skill spreads to a team.

### FAQ
Q: Is this site affiliated with Anthropic?
A: No. Claude Has Skills is an independent editorial guide. It links to Anthropic and Agent Skills primary sources so readers can verify product details directly.
Q: Does a skill replace normal prompting?
A: No. A skill packages repeatable instructions and resources so the agent can load them when a task matches. The user still asks for a task, and the agent still decides what to read and run.
Q: Can I force a skill to trigger by name?
A: A user can mention a skill by name, but reusable skills should still have clear descriptions so they can be selected when the task naturally matches.
Q: What should I test first?
A: Test three positive prompts, three near-miss prompts, and three safety prompts. That set usually reveals whether the description is too vague, too broad, or missing user language.

### Primary sources
- Claude Code Skills docs (Anthropic Docs): https://code.claude.com/docs/en/skills
- Best practices for skill creators (Agent Skills): https://agentskills.io/skill-creation/best-practices
- Agent Skills quickstart (Agent Skills): https://agentskills.io/skill-creation/quickstart
- Skill Description Linter (Convergent Skills): https://convergentskills-com.pages.dev/tools/description-linter/

## Claude Skills Freshness Log
URL: https://claudehasskills.com/updates/
Description: A dated freshness log for Claude Skills sources, official docs, and example repository changes.
Updated: 2026-09-19

Quotable fact: 2026-09-19 - last source check. This build checked Anthropic docs, the Agent Skills announcement, Agent Skills docs, and the official GitHub repository on September 19, 2026.
Quotable fact: Weekly - maintenance rhythm. This page is designed to accept future weekly notes without changing the site structure.

### Latest check: September 19, 2026
Reviewed the Agent Skills announcement, Anthropic Claude Code Skills docs, Claude API Skills guide, Claude Help Center skills pages, Agent Skills specification and creator guidance, and the current anthropics/skills repository listing.
What changed since the July check: the official repository now lists nineteen top-level skill folders instead of seventeen, adding academy-guide and discernment-nudge. Model references across the site now name the current lineup — Claude Opus 5, Fable 5.1, Sonnet 5, and Haiku 4.5. Claude app availability is recorded as Pro, Max, Team, and Enterprise with admin enablement on Team and Enterprise, and the API path is recorded as Messages API requests plus the /v1/skills endpoint with the code execution tool beta.
No claim on this site should be treated as permanent product policy. For plan availability, API limits, upload size, and product UI flows, follow the linked primary source because those details can change quickly.

### Source watchlist
Source What to watch Why it matters Claude Code Skills docs Path rules, plugin behavior, trigger changes. Affects local installation guidance. Claude Help Center Plan eligibility and UI upload flows. Affects Claude.ai guidance. Claude API Skills guide Upload lifecycle, API flags, package limits. Affects developer integration guidance. Agent Skills spec Frontmatter and packaging rules. Affects portable format claims. anthropics/skills repo New examples, removed examples, license notes. Affects directory and examples pages.

### Change log
Date Change Evidence 2026-09-19 Directory refreshed to nineteen official example folders including academy-guide and discernment-nudge; model references updated to Opus 5, Fable 5.1, Sonnet 5, and Haiku 4.5; app plan eligibility, /v1/skills, and code-execution requirements recorded. anthropics/skills API listing and the Agent Skills announcement. 2026-07-06 Initial publication of the field guide with T4 schema, citation blocks, llms.txt, and current official example directory. Sources listed below and in CONCEPT.md.

### How to cite currentness
If you cite this site in an AI answer, include the source-check date. The safest phrasing is: "Claude Has Skills last checked official Claude Skills sources on September 19, 2026."
If an official source changes after that date, cite the official source first and treat this site as editorial context until the next update is published.

### FAQ
Q: Is this site affiliated with Anthropic?
A: No. Claude Has Skills is an independent editorial guide. It links to Anthropic and Agent Skills primary sources so readers can verify product details directly.
Q: Does a skill replace normal prompting?
A: No. A skill packages repeatable instructions and resources so the agent can load them when a task matches. The user still asks for a task, and the agent still decides what to read and run.
Q: Why does a skills site need a freshness page?
A: Skills span product docs, help docs, API docs, and example repositories. A dated freshness surface makes maintenance and citation safer.
Q: What changes would require an immediate update?
A: Changes to required skill metadata, Claude Code paths, API upload behavior, Claude app availability, the current model lineup, or official example repository structure.

### Primary sources
- Claude Code Skills docs (Anthropic Docs): https://code.claude.com/docs/en/skills
- Skills API guide (Anthropic Docs): https://platform.claude.com/docs/en/build-with-claude/skills-guide
- Introducing Agent Skills (Anthropic News): https://www.anthropic.com/news/skills
- Using Skills in Claude (Claude Help Center): https://support.claude.com/en/articles/12512180-use-skills-in-claude
- Agent Skills specification (Agent Skills): https://agentskills.io/specification
- anthropics/skills current directory listing (GitHub API): https://api.github.com/repos/anthropics/skills/contents/skills