Anthropic Agent Skills
Anthropic's public Agent Skills repository for Claude, with example skills, document skills, the Agent Skills specification pointer, a template skill, Claude Code plugin installation, Claude.ai usage guidance, and Claude API skill creation references.
Open the source and read safety notes before installing.
Safety notes
- The repository is explicitly presented as demonstration and educational material; behavior available in Claude products can differ from repository examples.
- Document skills can read, transform, and create files such as PDFs, Word documents, PowerPoint decks, and spreadsheets, so review generated artifacts before sharing or publishing them.
- The MCP builder skill guides agents through creating MCP servers and evaluations; generated servers still need security review, least-privilege tool design, authentication review, and transport testing.
- The skill creator workflow can generate instructions, scripts, references, and assets; review all generated resources before installing them into an agent host.
Privacy notes
- Skill usage can expose documents, spreadsheets, slide decks, prompts, company workflows, brand guidelines, MCP designs, API details, and generated artifacts to Claude or the configured model/runtime.
- Do not upload or commit customer documents, regulated data, private brand assets, secrets, credentials, or unreleased business plans unless the environment and account policy allow that data.
- When creating custom skills, check bundled scripts, references, and assets for private data before sharing a repository, plugin, or ZIP archive.
Prerequisites
- Claude Code for plugin marketplace installation, Claude.ai paid plan access for built-in/example skill usage, or Claude API access for programmatic skill creation.
- A target task where a demonstration skill, document skill, MCP builder workflow, skill creator workflow, or custom skill template is appropriate.
- Review of the repository disclaimer and license/source-availability notes before relying on a skill in production.
- Local testing for any copied or customized skill before using it on sensitive documents, production workflows, or customer data.
Schema details
- Install type
- package
- Reading time
- 7 min
- Difficulty score
- 72
- Troubleshooting
- Yes
- Breaking changes
- No
- Scope
- Source repo
- Skill type
- capability-pack
- Skill level
- expert
- Verification
- validated
- Verified at
- 2026-06-18
| Platform | Support | Install path |
|---|---|---|
| claude-code | Native | .claude/skills/<skill-name>/SKILL.md |
| codex | Native | .agents/skills/<skill-name>/SKILL.md |
| windsurf | Native | .windsurf/skills/<skill-name>/SKILL.md |
| gemini | Native | .gemini/skills/<skill-name>/SKILL.md or .agents/skills/<skill-name>/SKILL.md |
| cursor | Adapter | .cursor/rules/<skill-name>.mdc |
| cli | Manual | AGENTS.md or tool-specific context file |
Full copyable content
/plugin marketplace add anthropics/skills
# Install a plugin from the marketplace
/plugin install document-skills@anthropic-agent-skills
/plugin install example-skills@anthropic-agent-skillsAbout this resource
Anthropic Agent Skills
anthropics/skills is Anthropic's public repository for Claude skills. It
contains example skills, document skills, an Agent Skills specification pointer,
a minimal template skill, and Claude Code plugin marketplace instructions.
Use this listing for Anthropic's own public skills repository and reference materials. Use narrower entries when evaluating a single community skill, third-party skill pack, MCP server, or tool.
Knowledge Freshness
Claude Code plugin behavior, Claude.ai skill availability, Claude API skill
endpoints, and the Agent Skills specification can change. Verify current
Anthropic support docs, Claude API docs, and the live
https://agentskills.io/specification page before building production
automation around repository examples.
The repository README says these skills demonstrate patterns and possibilities, and that implementations or behaviors in Claude products may differ. Treat the repository as a source-backed reference, not a guarantee that every listed skill is available or identical in every Claude surface.
Retrieval Sources
This listing is grounded in:
- The upstream
anthropics/skillsREADME. - Representative
SKILL.mdfiles formcp-builderandskill-creator. - The template
SKILL.mdfor the minimal required skill structure. - The repository spec pointer to the Agent Skills specification.
- Current GitHub repository metadata.
Core Workflow
Register the repository as a Claude Code plugin marketplace:
/plugin marketplace add anthropics/skills
Install a plugin from that marketplace:
/plugin install document-skills@anthropic-agent-skills
/plugin install example-skills@anthropic-agent-skills
After installation, invoke a skill by task. For example:
Use the PDF skill to extract the form fields from path/to/file.pdf.
Claude.ai users can use built-in/example skills where available on paid plans, and Claude API users can use pre-built or custom skills through Anthropic's Skills API workflow.
Capability Scope
The README organizes the repository into:
| Area | Scope |
|---|---|
skills/ |
Creative and design examples, development and technical examples, enterprise and communication examples, and document skills |
spec/ |
Pointer to the Agent Skills specification |
template/ |
Minimal starter SKILL.md structure |
Skill examples include document formats such as docx, pdf, pptx, and
xlsx, plus development-oriented examples such as mcp-builder,
skill-creator, webapp-testing, and claude-api.
The README states that many skills are open source under Apache 2.0, while the
document creation and editing skills under docx, pdf, pptx, and xlsx
are source-available references rather than open source.
Production Rules
- Treat this repository as a reference and demo collection unless a specific Claude surface documents production support for the exact skill.
- Test skills thoroughly in your own environment before relying on them for critical workflows.
- Review document outputs, generated files, and transformed data before sharing or publishing.
- Review generated MCP servers for authentication, tool permissions, transport choices, error handling, and evaluation coverage before deployment.
- Do not copy example skills into internal plugins without checking license, source-availability, data-handling, and security requirements.
- Keep customer documents, secrets, regulated data, and proprietary workflows out of public examples, issues, screenshots, and committed custom skills.
Use Cases
- Install Anthropic's document skills in Claude Code for PDF, DOCX, PPTX, or XLSX workflows where available.
- Use
mcp-builderas a reference workflow for planning, implementing, and evaluating a high-quality MCP server. - Use
skill-creatorto understand the structure, iteration loop, and testing approach for custom skills. - Browse the template skill when creating a minimal
SKILL.md. - Use the repository to learn how Claude skills package instructions, scripts, references, and assets.
Source Review
Verified on 2026-06-18:
- The README describes skills as folders of instructions, scripts, and resources that Claude loads dynamically for specialized tasks.
- The README says the repository contains Anthropic's implementation of skills
for Claude and points the Agent Skills standard to
agentskills.io. - The README documents Claude Code plugin marketplace installation with
/plugin marketplace add anthropics/skillsand direct installs fordocument-skillsandexample-skills. - The README describes Claude.ai paid-plan usage and Claude API skill creation through Anthropic docs.
- The
mcp-builderskill describes MCP server design, tool naming, context management, SDK choices, testing with MCP Inspector, and evaluation creation. - The
skill-creatorskill describes drafting, testing, evaluating, and iterating custom skills withSKILL.md, scripts, references, and assets. - The template skill shows the minimal
nameanddescriptionfrontmatter. - The repository README distinguishes open-source example skills from source-available document skills.
Duplicate Review
Checked current content/skills/, content/tools/, content/mcp/,
content/agents/, open pull requests, and repository-wide content for
anthropics/skills, Anthropic Agent Skills, Claude Code skills, Claude API
skills, mcp-builder, skill-creator, document skills, and Agent Skills
specification. Existing content discusses Anthropic skills generally and covers
some skill-authoring concepts, but no dedicated anthropics/skills repository
entry, exact source URL duplicate, target file, or open duplicate PR was found.
Disclosure
Editorial listing. No paid placement or affiliate link is used. The repository is maintained by Anthropic and includes mixed open-source and source-available skill materials as described in the upstream README.
Source citations
Add this badge to your README
How it compares
Anthropic Agent Skills side by side with 3 alternatives on trust, install, platform support, and disclosed safety notes — all from reviewed registry metadata.
| Field | Anthropic Agent Skills Anthropic's public Agent Skills repository for Claude, with example skills, document skills, the Agent Skills specification pointer, a template skill, Claude Code plugin installation, Claude.ai usage guidance, and Claude API skill creation references. Open dossier | Google Cloud Agent Skills Google Agent Skills catalog for AI agents working with Google Cloud, Gemini Enterprise Agent Platform, Gemini APIs, Skill Registry, Cloud Run, BigQuery, Firebase, GKE, Cloud SQL, AlloyDB, gcloud, auth, onboarding, and Well-Architected Framework guidance. Open dossier | mcp-use MCP Apps Builder Skill Agent Skill from mcp-use for building production MCP servers and MCP Apps with tools, resources, prompts, widgets, authentication, deployment, inspector testing, and framework-specific guardrails. Open dossier | Dart Agent Skills Official Dart team Agent Skills for AI coding agents working on Dart unit tests, CLI apps, coverage, runtime errors, mocks, package conflicts, static analysis, Native Assets, FFI, ffigen, and pattern matching. Open dossier |
|---|---|---|---|---|
| Trust | ||||
| Install risk | Review first | Review first | Review first | Review first |
| Notes | Safety ✓ Privacy ✓ | Safety ✓ Privacy ✓ | Safety ✓ Privacy ✓ | Safety ✓ Privacy ✓ |
| Category | skills | skills | skills | skills |
| Source | source-backed | source-backed | source-backed | source-backed |
| Author | Anthropic | mcp-use | Dart Team | |
| Added | 2026-06-18 | 2026-06-18 | 2026-06-18 | 2026-06-18 |
| Platforms | Claude CodeCodexWindsurfGeminiCursorCLI | Claude CodeCodexWindsurfGeminiCursorCLI | Claude CodeCodexWindsurfGeminiCursorCLI | Claude CodeCodexWindsurfGeminiCursorCLI |
| Source repo | — | — | — | — |
| Safety notes | ✓The repository is explicitly presented as demonstration and educational material; behavior available in Claude products can differ from repository examples. Document skills can read, transform, and create files such as PDFs, Word documents, PowerPoint decks, and spreadsheets, so review generated artifacts before sharing or publishing them. The MCP builder skill guides agents through creating MCP servers and evaluations; generated servers still need security review, least-privilege tool design, authentication review, and transport testing. The skill creator workflow can generate instructions, scripts, references, and assets; review all generated resources before installing them into an agent host. | ✓Google Cloud skills can create, update, delete, deploy, query, or configure cloud resources, datasets, IAM policies, service accounts, APIs, containers, jobs, and networking. The gcloud skill requires command validation and safety guardrails before invoking Google Cloud CLI commands; do not let agents improvise cloud commands from memory. Skill Registry guidance includes skill lifecycle management such as upload, update, and permanent delete operations; validate environment and approval before use. Cloud Run, GKE, BigQuery, Firebase, Cloud SQL, AlloyDB, and Gemini API workflows can create cost, expose endpoints, alter data, or change production behavior. The repository notes active development, so verify exact commands, product names, API availability, and launch-stage limits before production use. | ✓This skill guides agents through MCP server work that can expose tools, resources, prompts, widgets, middleware, auth flows, and remote deployment surfaces. MCP tools can call external APIs, read or mutate local data, proxy other MCP servers, or trigger side effects depending on implementation. Widget and MCP Apps work can render user-provided data in interactive UI surfaces; validate data and avoid trusting client-side state. Authentication examples may involve OAuth, Supabase, Clerk, Auth0, WorkOS, Keycloak, custom auth, API keys, and session handling. Keep secrets in environment variables. Deployment guidance can make a server reachable by external clients; review tool authority, CORS, auth, logging, rate limits, and observability before publishing. | ✓Dart skills can modify source code, tests, pubspec dependencies, analysis options, generated mocks, coverage output, native hooks, FFI bindings, and package constraints. `dart fix --apply`, formatter runs, mock generation, build_runner, ffigen, and Native Assets hooks can create broad mechanical diffs; inspect changes before committing. Native Assets and FFI workflows may compile C/C++ code, run local toolchains, download precompiled binaries, and link dynamic or static libraries. Downloaded binaries should use cryptographic hash validation and offline fallbacks as the skill describes. Package conflict fixes can upgrade or downgrade dependencies and affect transitive package behavior across Dart and Flutter projects. |
| Privacy notes | ✓Skill usage can expose documents, spreadsheets, slide decks, prompts, company workflows, brand guidelines, MCP designs, API details, and generated artifacts to Claude or the configured model/runtime. Do not upload or commit customer documents, regulated data, private brand assets, secrets, credentials, or unreleased business plans unless the environment and account policy allow that data. When creating custom skills, check bundled scripts, references, and assets for private data before sharing a repository, plugin, or ZIP archive. | ✓Google Cloud workflows may expose project IDs, service account emails, OAuth tokens, API keys, ADC credentials, Terraform state, dataset names, table schemas, query text, logs, traces, prompts, model outputs, embeddings, and customer data. BigQuery, Firebase, Cloud SQL, AlloyDB, GKE, and Agent Platform workflows may process regulated or proprietary data; review data residency, IAM, retention, audit logging, and sharing rules before use. Gemini API and Agent Platform skills can send prompts, files, images, audio, video, tool inputs, structured outputs, cached contexts, and batch datasets to Google services. Keep credentials, project IDs when sensitive, private queries, logs, trace payloads, Terraform state, customer data, and generated datasets out of public prompts, issues, PRs, and screenshots. | ✓MCP Apps and servers may process prompts, tool arguments, API responses, user identity, OAuth scopes, logs, widget props, resources, and generated UI state. The skill's reference files can lead agents to inspect local source, configs, package metadata, auth providers, deployment logs, and connected MCP server definitions. Do not paste API keys, OAuth secrets, cookies, session tokens, production customer data, or private server URLs into prompts, examples, generated docs, issues, or PRs. When using hosted deployment or inspector tools, review the retention policies for mcp-use, Manufact, MCP clients, model providers, and connected services. | ✓Dart projects may contain private package names, API schemas, service URLs, test fixtures, generated mocks, coverage paths, stack traces, native library names, and build logs. Runtime-error workflows can expose stack traces, local paths, active app state, logs, and user-input data from the running program. FFI and Native Assets workflows may include proprietary C/C++ source, platform-specific build settings, binary hashes, and downloaded artifact URLs. Keep credentials, private package registries, source paths when sensitive, proprietary native code, test data, and generated coverage reports out of public prompts, issues, PRs, and examples. |
| Prerequisites |
|
|
|
|
| Install | | | | |
| Config | — | — | — | — |
| Citations | ||||
| Claim | Unclaimed | Unclaimed | Unclaimed | Unclaimed |
Signals
Loading live community signals…
A short, calm digest of reviewed Claude resources. Unsubscribe any time.