Independent reviews. IonBin has no partnerships with the tools on this page and was not paid for these listings. Our evaluations are our own, based on public documentation and our research. Product names are trademarks of their owners and are used for identification only. How disclosure works.

The agent skills field guide

Skills.md

Know what you’re adding
to your agent.

Find useful skills. Understand the setup, the access they need, and the things to check before you run them.

Explore the skills How we evaluate
43skills in the collection
21task categories
07records with runtime evidence

Collection updated

A practical starting point

Find the right fit.

Requirements first. Evidence attached.
Popularity is a discovery signal, not a recommendation.

Evidence in view

20 source-only assessments · 7 runtime probes · 16 decision scenarios. Review coverage and our experience with a skill are separate. Runtime probes test bounded behavior; task results apply to their recorded conditions. Neither is a universal security verdict.

Frontend

Vercel

React Best Practices

A prioritized React and Next.js review guide covering request waterfalls, bundle size, rendering and server behavior.

What you’ll need

  • An agent able to read SKILL.md and the referenced rule files; no executable helper is invoked by the reviewed entrypoint.
  • Access to the target React/Next.js source. Building and measuring changes needs that project's own runtime and tooling.
Check before use

Server changes need security review

Scenario checkedExplored · Oct 18, 2026

7 passed · 0 failed · 7 totalRecorded scope completed · Evidence report ↗

Read assessment GitHub
Design & accessibility

Vercel

Web Design Guidelines

Reviews UI source against Vercel's interface rules and returns findings with file and line references.

What you’ll need

  • An agent with file-reading and an HTTP-fetch capability; the entrypoint names WebFetch.
  • Network access to raw.githubusercontent.com for the external guideline file on every review.
Check before use

Mutable remote instructions

Scenario checkedExplored · Oct 18, 2026

8 passed · 0 failed · 8 totalRecorded scope completed · Evidence report ↗

Read assessment GitHub
Skill authoring

Anthropic

Skill Creator

Drafts and improves skills through example tasks, baseline comparisons, human feedback and trigger evaluations.

What you’ll need

  • Drafting needs an agent and writable workspace; automated trigger evaluation additionally uses Python 3 and the claude CLI.
  • The description improver uses the session's Claude Code authentication; its source explicitly says no separate ANTHROPIC_API_KEY is needed.
Check before use

Evaluation text can escape the viewer script context

Runtime probedExplored · Oct 18, 2026

5 passed · 2 failed · 7 totalRecorded scope completed · Evidence report ↗

Read assessment GitHub
Documents

Anthropic

PDF

Guidance and helpers for PDF extraction, creation, page operations, OCR and form filling.

What you’ll need

  • Python with task-specific libraries: pypdf for page/form operations, pdfplumber for extraction and ReportLab for creation.
  • Image/OCR examples use pdf2image and pytesseract; native Poppler/Tesseract setup must also be checked in the target environment. CLI examples include qpdf, Poppler tools and optional pdftk.
Check before use

Non-form PDF raises AttributeError

Runtime probedExplored · Oct 18, 2026

7 passed · 2 failed · 9 totalRecorded scope completed · Evidence report ↗

Read assessment GitHub
Documents

Anthropic

DOCX

Creates and edits Word documents through JavaScript generation, XML editing and rendered-output checks.

What you’ll need

  • Creation examples use Node.js and the docx npm package; reading examples use Pandoc.
  • Editing and verification use Python, ZIP tools, LibreOffice (soffice) and Poppler's pdftoppm; merge_runs.py imports defusedxml.
Check before use

Restricted reuse terms

Runtime probedExplored · Oct 18, 2026

7 passed · 0 failed · 7 totalRecorded scope completed · Evidence report ↗

Read assessment GitHub
Engineering

Jesse Vincent / Superpowers

Systematic Debugging

A four-stage debugging method: investigate evidence, compare patterns, test a hypothesis and verify one focused fix.

What you’ll need

  • A coding agent with access to relevant source, logs and the project's reproduction/test environment.
  • The optional find-polluter.sh helper needs Bash, Unix utilities and npm test; the core method is not restricted to JavaScript.
Check before use

Diagnostic data needs boundaries

Scenario checkedExplored · Oct 18, 2026

8 passed · 0 failed · 8 totalRecorded scope completed · Evidence report ↗

Read assessment GitHub
Engineering

Jesse Vincent / Superpowers

Test-Driven Development

Guides an agent through a failing behavior test, the smallest passing change and subsequent refactoring.

What you’ll need

  • A coding agent able to edit tests and implementation and run the target project's test suite.
  • A runnable test framework for the language in use. npm test and TypeScript are examples, not a universal runtime requirement.
Check before use

Delete-and-restart instruction

Scenario checkedExplored · Oct 18, 2026

8 passed · 0 failed · 8 totalRecorded scope completed · Evidence report ↗

Read assessment GitHub
Browser automation

OpenAI catalog / Microsoft CLI

Playwright CLI

A terminal-driven browser workflow for navigation, form interaction, screenshots and UI debugging. This catalog version is a historical reference.

What you’ll need

  • Node.js/npm with npx on PATH, Bash for the bundled wrapper and a working browser runtime.
  • The wrapper resolves @playwright/cli through npx; package/browser setup and target-page access may require networking.
Check before use

Unpinned executable dependency

Runtime probedExplored · Oct 18, 2026

6 passed · 0 failed · 6 totalRecorded scope completed · Evidence report ↗

Read assessment GitHub
Skill discovery

Vercel

Find Skills

A discovery workflow for locating and installing agent skills through the Skills CLI.

What you’ll need

  • The reviewed Skills CLI package declares Node.js >=22.20.0; npm/npx and Git/network access are needed for its documented installation flow.
  • No service account or API key is specified for public discovery. Private repositories need their own authorized access.
Check before use

Global, unattended installs

Scenario checkedExplored · Oct 18, 2026

8 passed · 0 failed · 8 totalRecorded scope completed · Evidence report ↗

Read assessment GitHub
Design & accessibility

Anthropic

Frontend Design

An art-direction workflow that turns a brief into deliberate typography, palette, layout, and responsive implementation.

What you’ll need

  • A coding agent with project-file access; the target application's own runtime/build tools for implementation.
  • A screenshot-capable browser is useful for visual critique. No dedicated package, account, or API key is required by the reviewed instruction file.
Check before use

Context can enter public output

Scenario checkedExplored · Oct 18, 2026

8 passed · 0 failed · 8 totalRecorded scope completed · Evidence report ↗

Read assessment GitHub
Browser automation

Anthropic

Webapp Testing

Python Playwright guidance for local web-app inspection, screenshots, and scripted UI checks.

What you’ll need

  • Python 3, the Python Playwright package and its Chromium browser for the documented examples.
  • The application must have its own installed dependencies and launch command; with_server.py itself uses Python standard-library modules.
Check before use

Shell authority and incomplete process isolation

Scenario checkedUsed · Oct 18, 2026

9 passed · 0 failed · 9 totalRecorded scope completed · Evidence report ↗

Read assessment GitHub
Browser automation

Vercel

Agent Browser

A browser-automation discovery skill backed by a native CLI and version-specific core guidance.

What you’ll need

  • The reviewed npm package is agent-browser 0.39.1 and declares Node.js >=24; installation obtains a platform-specific native binary.
  • Chrome/Chromium via CDP is the main browser path. The source provides Windows/macOS/Linux handling; pnpm >=11 is declared for package development.
Check before use

Browser session authority

Scenario checkedExplored · Oct 18, 2026

8 passed · 0 failed · 8 totalRecorded scope completed · Evidence report ↗

Read assessment GitHub
Video & motion

Remotion

Remotion Best Practices

A router to Remotion guidance for creating, previewing and rendering programmatic videos.

What you’ll need

  • Node.js and Git for the reviewed project-creation flow; npm dependencies include Remotion and React.
  • A browser for Studio preview; rendering requires the Remotion runtime and its platform-specific dependencies.
Check before use

Mutable executable setup

Scenario checkedExplored · Oct 18, 2026

8 passed · 0 failed · 8 totalRecorded scope completed · Evidence report ↗

Read assessment GitHub
Frontend

Vercel

React Composition Patterns

Guidance for compound components, shared state and explicit component variants.

What you’ll need

  • A React project and coding agent; no dedicated executable helper, account or API key is specified.
  • The React 19 API section applies only to React 19+. React 18 projects must skip those changes.
Check before use

Version-sensitive refactors

Scenario checkedExplored · Oct 18, 2026

8 passed · 0 failed · 8 totalRecorded scope completed · Evidence report ↗

Read assessment GitHub
Marketing

Corey Haines

SEO Audit

A structured audit framework for crawlability, technical foundations, on-page content and evidence-based priorities.

What you’ll need

  • A browser or fetch tool and permission to inspect target URLs. Rendered-page checks are needed for client-injected schema.
  • Search Console/analytics access is needed for property-specific performance evidence; public-page checks can proceed without it.
Check before use

Private marketing and analytics context

Scenario checkedExplored · Oct 18, 2026

9 passed · 0 failed · 9 totalRecorded scope completed · Evidence report ↗

Read assessment GitHub
Marketing

Corey Haines

Copywriting

A brief-first workflow for preserving voice, tightening scope and drafting supportable marketing claims.

What you’ll need

  • A writing agent, an approved brief and support for factual claims.
  • Relevant product-marketing context may be read from local files; no dedicated runtime, paid tool, account or API key is required.
Check before use

Confidential customer evidence

Scenario checkedExplored · Oct 18, 2026

8 passed · 0 failed · 8 totalRecorded scope completed · Evidence report ↗

Read assessment GitHub
Engineering

Jesse Vincent / obra

Requesting Code Review

A workflow for delegating a bounded code review with requirements and an explicit Git range.

What you’ll need

  • Git history with meaningful base/head commits, a review-capable subagent and the reviewer template.
  • No dedicated package or API key is specified. Host model access and project test tools remain separate.
Check before use

Wrong range omits the work

Scenario checkedExplored · Oct 18, 2026

9 passed · 0 failed · 9 totalRecorded scope completed · Evidence report ↗

Read assessment GitHub
Security review

Trail of Bits

Differential Security Review

A risk-focused review method connecting code changes to history, affected callers, test coverage and explicit reports.

What you’ll need

  • A local Git repository, baseline/head revisions, source-reading/search tools and report-file access.
  • The PR path uses gh and authorized GitHub access. Local commit review does not require an account or API key.
Check before use

Checkout can disturb active work

Scenario checkedExplored · Oct 18, 2026

9 passed · 0 failed · 9 totalRecorded scope completed · Evidence report ↗

Read assessment GitHub
Design & accessibility

Next Level Builder

UI/UX Pro Max

A searchable local design-guidance library with optional persistent design-system output.

What you’ll need

  • Python 3.x; reviewed search/core modules use standard-library imports and local CSV data. Keep the script/data layout intact.
  • Resolve the full installed script path; the documented CLAUDE_PLUGIN_ROOT path may need adaptation outside that plugin host.
Check before use

Persistence and force overwrite

Scenario checkedExplored · Oct 18, 2026

8 passed · 0 failed · 8 totalRecorded scope completed · Evidence report ↗

Read assessment GitHub
Design & accessibility

leonxlnx

Taste Frontend Design

An opinionated design workflow using direction selection, browser inspection, screenshot critique and asset sourcing.

What you’ll need

  • Full inspector workflow: Node.js 22+ and installed Chrome, Edge or Chromium. Browser discovery covers Windows, macOS and Linux paths.
  • The lint fallback and assets helper state Node 18+; no npm dependencies are required by the selected helpers.
Check before use

Local serving and sensitive output

Scenario checkedExplored · Oct 18, 2026

8 passed · 0 failed · 8 totalRecorded scope completed · Evidence report ↗

Read assessment GitHub
Documents

Anthropic

PowerPoint Presentations

A presentation authoring and editing workflow with OOXML helpers, template guidance and rendered-slide review. Useful for editable decks, with substantial file-mutation and licensing caveats.

What you’ll need

  • A host agent with file and shell access in an environment where the per-skill terms permit use. The reviewed license is proprietary, not the repository's Apache-licensed skill terms.
  • Node and pptxgenjs for new decks; Python and named XML/image packages for helpers; markitdown for text extraction; LibreOffice and PDF image tools for rendering. Upstream claims of preinstallation do not establish availability on your machine.
Check before use

Runtime not run: restrictive per-skill terms

Source reviewedExplored · Oct 19, 2026

4 passed · 2 failed · 6 totalRecorded scope completed · Evidence report ↗

Read assessment GitHub
Documents

Anthropic

Excel Spreadsheets

A spreadsheet workflow for formula-based workbooks, formatting and recalculation checks. It makes useful distinctions between formulas and cached values, but verification can rewrite the workbook and does not prove the calculations are correct.

What you’ll need

  • An agent with local file and shell access in an environment permitted by the proprietary per-skill terms; the adjacent Apache-licensed skills do not extend their license to this folder.
  • Python with openpyxl, pandas and markitdown as needed; LibreOffice/soffice for actual formula recalculation. Formula strings alone do not supply cached results.
Check before use

Runtime not run: restrictive per-skill terms

Source reviewedExplored · Oct 19, 2026

5 passed · 1 failed · 6 totalRecorded scope completed · Evidence report ↗

Read assessment GitHub
Development

Anthropic

MCP Server Builder

A guide for building MCP servers with clear tool contracts, transport choices and evaluation questions. Its included evaluation runner needs extra care around parallel tool calls and diagnostic disclosure.

What you’ll need

  • An agent able to read the skill and references; Node/TypeScript or Python and an appropriate MCP SDK for actual server development.
  • Access to the target API specification and a deliberately scoped test service. Real integrations may require service credentials and network access.
Check before use

Multiple tool requests are only partly handled

Runtime probedExplored · Oct 19, 2026

3 passed · 3 failed · 6 totalRecorded scope completed · Evidence report ↗

Read assessment GitHub
Design

Anthropic

Slack GIF Creator

A Python animation toolkit with drawing helpers, frame assembly and a lightweight validator. Useful for small custom animations; the validator's ready result covers dimensions rather than complete upload readiness.

What you’ll need

  • Python plus Pillow and numpy for the exercised frame and validator helpers; imageio for GIFBuilder.save. The requirements file also lists imageio-ffmpeg.
  • An agent with local file/code execution access and an output folder. No Slack account is needed to create or inspect local files; uploading requires a workspace and appropriate permission.
Check before use

The ready result checks dimensions, not full readiness

Runtime probedExplored · Oct 19, 2026

4 passed · 4 failed · 8 totalRecorded scope completed · Evidence report ↗

Read assessment GitHub
Planning

Matt Pocock

Grilling

A decision interview that groups questions by dependency and makes unresolved assumptions visible before work starts.

What you’ll need

  • An agent that can load the instruction file and retain the interview context.
  • A user available to settle consequential decisions. Subagent and filesystem/search tools are needed only for delegated fact finding.
Check before use

Fact finding can widen tool access

Source reviewedExplored · Oct 19, 2026

5 passed · 0 failed · 5 totalRecorded scope completed · Evidence report ↗

Read assessment GitHub
Agent workflows

Matt Pocock

Handoff

A compact continuation note that points to existing artifacts, records the next focus and asks the agent to redact sensitive content.

What you’ll need

  • A host that exposes the current conversation and can write a text file.
  • Access to the existing artifacts the handoff references; receiving agent access is a separate requirement.
Check before use

Redaction is an instruction, not a control

Source reviewedExplored · Oct 19, 2026

5 passed · 0 failed · 5 totalRecorded scope completed · Evidence report ↗

Read assessment GitHub
Databases

Supabase

Supabase Postgres Best Practices

A topic-based SQL review guide covering query plans, schema design, connection management and row-level security.

What you’ll need

  • An agent able to read the topic files and the relevant SQL or schema.
  • A separately configured disposable Postgres database is needed for runtime validation; Supabase-specific auth.uid() examples need matching identity facilities.
Check before use

A tenant setting is not authentication

Source reviewedExplored · Oct 19, 2026

6 passed · 0 failed · 6 totalRecorded scope completed · Evidence report ↗

Read assessment GitHub
Marketing

Corey Haines

Conversion Rate Optimization

A structured review of marketing pages and lead forms, with prioritized fixes, copy alternatives and experiments to validate.

What you’ll need

  • An agent with page copy or a permitted page snapshot and product/audience context.
  • Product-marketing context files are optional but explicitly checked. Signup and test-design tasks route to separate skills.
Check before use

Recommendations are test hypotheses

Source reviewedExplored · Oct 19, 2026

6 passed · 0 failed · 6 totalRecorded scope completed · Evidence report ↗

Read assessment GitHub
Developer safety

Matt Pocock

Git Guardrails for Claude Code

A Bash hook setup guide that blocks selected Git command text, with important gaps in command parsing and error handling.

What you’ll need

  • Claude Code hook support and an operator-chosen project or global settings scope.
  • Bash, jq, grep and cat must be available to the hook. Git Bash could not start in this evaluation sandbox.
Check before use

Substring patterns are not a Git permission boundary

Source reviewedExplored · Oct 19, 2026

6 passed · 0 failed · 6 totalRecorded scope completed · Evidence report ↗

Read assessment GitHub
Planning

Superpowers

Brainstorming

A requirements conversation with explicit scope choices, a design review and an optional local visual companion.

What you’ll need

  • An instruction-capable agent and a user who can confirm design decisions. Subagents are optional for the builder check.
  • Optional visual companion: Node.js, a browser, a local listening port and Bash for the supplied launcher scripts. The text-only path needs none of those helper services.
Check before use

Local previews still contain active web content

Runtime probedExplored · Oct 19, 2026

7 passed · 0 failed · 7 totalRecorded scope completed · Evidence report ↗

Read assessment GitHub
Writing & comms

Anthropic

Internal Comms

A small router that picks a format guide for status reports, newsletters, FAQs and other internal updates.

What you’ll need

  • An agent that can load the skill folder, including its examples/ guide files.
  • No account or script is needed. Source material for the update comes from you.
Check before use

The format guides are examples, not your house style

Source reviewedExplored · Oct 19, 2026

4 passed · 0 failed · 4 totalRecorded scope completed · Evidence report ↗

Read assessment GitHub
Meetings & notes

Notion (in the OpenAI skills catalog)

Notion Meeting Intelligence

Meeting prep that pulls context from Notion over MCP and drafts an agenda or pre-read page with owners and timeboxes.

What you’ll need

  • Codex with the Notion MCP server connected over OAuth (the skill gives Codex-specific setup commands).
  • A Notion workspace with the relevant pages shared with the connection.
Check before use

It writes to Notion

Source reviewedExplored · Oct 19, 2026

4 passed · 0 failed · 4 totalRecorded scope completed · Evidence report ↗

Read assessment GitHub
Project management

OpenAI

Linear

A step-by-step routine for working in Linear through its MCP server: read first, then create or update, then summarize.

What you’ll need

  • A Linear workspace and the Linear MCP server connected over OAuth. The setup steps are Codex-specific.
  • Access to the teams and projects involved.
Check before use

Bulk changes can touch many issues

Source reviewedExplored · Oct 19, 2026

4 passed · 0 failed · 4 totalRecorded scope completed · Evidence report ↗

Read assessment GitHub
Meetings & notes

OpenAI

Transcribe

Transcribes audio through OpenAI's API with a bundled Python script, with optional speaker labels.

What you’ll need

  • Python with the openai package, and an OPENAI_API_KEY in the environment.
  • Audio files on the machine the agent can reach.
Check before use

Recordings leave your machine

Source reviewedExplored · Oct 19, 2026

4 passed · 0 failed · 4 totalRecorded scope completed · Evidence report ↗

Read assessment GitHub
Security review

OpenAI

Security Threat Model

Writes a repository-specific threat model: trust boundaries, assets, abuse paths and mitigations tied to evidence in the code.

What you’ll need

  • An agent with read access to the repository, and a person who can answer questions about deployment and users.
  • No account or service is needed to read the code.
Check before use

Not a security assurance

Source reviewedExplored · Oct 19, 2026

4 passed · 0 failed · 4 totalRecorded scope completed · Evidence report ↗

Read assessment GitHub
Advertising & analytics

Google

Google Ads Account Diagnostics

Diagnoses drops in Google Ads conversions, leads and impression share with GAQL queries over Google's read-only Ads MCP server.

What you’ll need

  • The Google Ads MCP server (search, list_accessible_customers, get_resource_metadata), configured with Google Ads API access.
  • Access to the client accounts being diagnosed.
Check before use

Recommendations still need judgment

Source reviewedExplored · Oct 19, 2026

5 passed · 0 failed · 5 totalRecorded scope completed · Evidence report ↗

Read assessment GitHub
Advertising & analytics

Google

Google Analytics Data API Basics

Sets up and runs Google Analytics 4 reports through the Data API v1beta, with read-only scopes and a field compatibility check.

What you’ll need

  • A Google Cloud project with the Analytics Data API enabled, the gcloud CLI, and access to the GA4 property.
  • A language runtime and client library (Python, Node.js, Java, PHP, Go, .NET or Ruby).
Check before use

Local credentials are broader than Analytics

Source reviewedExplored · Oct 19, 2026

5 passed · 0 failed · 5 totalRecorded scope completed · Evidence report ↗

Read assessment GitHub
Engineering

Cloudflare

Web Performance Audit

A five-phase page-speed audit using Chrome DevTools MCP traces, network analysis, an accessibility snapshot and a codebase check.

What you’ll need

  • An agent with browser tooling, ideally Chrome DevTools MCP, for traces and network data.
  • Codebase access for the optional fifth phase.
Check before use

Lab numbers are not user experience

Source reviewedExplored · Oct 19, 2026

4 passed · 0 failed · 4 totalRecorded scope completed · Evidence report ↗

Read assessment GitHub
Evaluation

Hugging Face

Hugging Face Community Evals

Runs benchmark evaluations of Hub models locally with inspect-ai or lighteval, starting from small smoke tests.

What you’ll need

  • uv, the bundled Python scripts, and an HF_TOKEN for gated or private models.
  • A local GPU for vLLM or Transformers runs, or Inference Providers for the lighter path.
Check before use

Remote code flag

Source reviewedExplored · Oct 19, 2026

4 passed · 0 failed · 4 totalRecorded scope completed · Evidence report ↗

Read assessment GitHub
Security review

Sentry

Skill Scanner

Reviews an agent skill before you install it, with a static scan script plus checks for injection, risky scripts, permissions and supply-chain links.

What you’ll need

  • uv for the bundled scanner script, and an agent with Read, Grep, Glob and Bash.
  • A local copy of the skill to scan.
Check before use

The scanner runs code on your machine

Source reviewedExplored · Oct 19, 2026

4 passed · 0 failed · 4 totalRecorded scope completed · Evidence report ↗

Read assessment GitHub
Evaluation

Sentry

Prompt Optimizer

Improves prompts against a small eval set: define success first, try a few candidate rewrites, keep the one that wins on held-out cases.

What you’ll need

  • An agent that can read the bundled reference files.
  • Example inputs and success criteria, or time to build a small eval set.
Check before use

Eval data can contain personal details

Source reviewedExplored · Oct 19, 2026

4 passed · 0 failed · 4 totalRecorded scope completed · Evidence report ↗

Read assessment GitHub
Databases

Supabase

Supabase

Supabase's general skill for agents: check current docs first, verify every fix, and follow a security checklist for auth, RLS, keys and storage.

What you’ll need

  • A Supabase project and either the Supabase CLI or the Supabase MCP server.
  • A scoped personal access token for CI or agents, per the skill.
Check before use

SQL runs with real privileges

Source reviewedExplored · Oct 19, 2026

5 passed · 0 failed · 5 totalRecorded scope completed · Evidence report ↗

Read assessment GitHub
Development

Microsoft

Microsoft Docs

Answers questions about Microsoft products from live Microsoft Learn documentation, through the Learn MCP server or the mslearn CLI.

What you’ll need

  • The Microsoft Learn MCP server at learn.microsoft.com/api/mcp, or Node.js for the mslearn CLI.
  • No Microsoft account is needed to search public documentation.
Check before use

The CLI fallback runs a downloaded package

Source reviewedExplored · Oct 19, 2026

4 passed · 0 failed · 4 totalRecorded scope completed · Evidence report ↗

Read assessment GitHub

The assessment is only useful if it’s clear.

See how the review works.

Our rubric separates source inspection, observed use and repeatable testing. It records the permissions, data flows, versions and unanswered questions.

Read the collection as Markdown or JSON. These are review records, not installable skills.

Read the methodology