Инструменты

Anthropic's document skills, the implementation behind Claude's file features: create, edit and analyze Word, Excel, PowerPoint and PDF files, including tracked changes, comments, formulas and form filling. Source-available, not under an open-source license.

Reviews pull requests with several parallel subagents, scores each finding by confidence to filter out false positives, and can post the results as PR comments. Official Anthropic plugin; requires a logged-in GitHub CLI.

Orchestrates subagents in-session to map the architecture, build a threat model and hunt for vulnerabilities. Every finding is independently verified before it goes in the report, which also comes as SARIF, and optional patch files are left for you to apply. Official Anthropic plugin.

A complete harness for coding agents: dozens of subagents, nearly 300 skills, hooks, rules, memory and continuous learning, plus the AgentShield security scanner. Built mainly for Claude Code, with Codex support.

Recreates the deprecated Explanatory output style with a SessionStart hook: adds Insight notes before and after code that explain implementation trade-offs and codebase conventions. Official Anthropic plugin; uses extra tokens.

Learning mode: at key decision points such as business logic or error handling, asks you to write 5-10 lines of code yourself while explaining the trade-offs. Official Anthropic plugin built on a SessionStart hook; uses extra tokens.

Six specialized review subagents that check comment accuracy, test coverage, silent failures, type design, code style and simplification opportunities. Run them individually or all at once. Official Anthropic plugin.

Addy Osmani's production-grade engineering skill pack covering spec, planning, implementation, testing, review and shipping, with commands like /spec, /plan, /build, /test, /review and /ship. Supports both Claude Code and Codex.

Official Chrome plugin bundling the Chrome DevTools MCP with debugging skills for accessibility, LCP and memory leaks: lets agents drive and inspect a real Chrome and analyze network requests, console logs and performance traces. Collects usage stats by default; can be disabled.

End-to-end academic paper plugin: deep research and literature review, writing, multi-perspective simulated peer review, revision and finalization, pausing for your approval at each stage, with built-in citation verification and integrity checks. Licensed CC BY-NC.

From Obsidian CEO Steph Ango (kepano): teaches agents to write Obsidian-flavored Markdown, Bases and Canvas files and work with vaults through the Obsidian CLI, plus skills like web page content extraction. Also installable into Codex and other agents via npx skills.

Browser automation plugin from Vercel Labs: agents drive Chrome through accessibility tree snapshots and element refs to navigate, click, fill forms and take screenshots, for web testing and exploratory QA. Install the agent-browser CLI first (npm, Homebrew or Cargo).

Multi-host plugin marketplace with 90+ single-purpose plugins, 200+ subagents and 180+ skills covering language-specific development, ops, security, data and multi-agent orchestration. Installing a plugin loads only that plugin's content.

Security plugin from the wshobson/agents marketplace: commands for SAST, dependency vulnerability scanning and hardening, security auditor and threat modeling subagents, and threat analysis skills such as STRIDE and attack trees.

Test-driven development plugin: /tdd-workflows:tdd-cycle and related commands move through red-green-refactor phases with a checkpoint for approval at each, plus TDD orchestrator and code reviewer subagents. Part of the wshobson/agents marketplace.

Orchestrates multi-agent teams using Claude Code's experimental Agent Teams feature: parallel multi-angle code review, competing-hypothesis debugging and feature work split by file ownership. Enable CLAUDE_CODE_EXPERIMENTAL_AGENT_TEAMS first.

Accessibility review plugin with an audit command, a UI validation subagent, and WCAG 2.2 and screen reader testing skills. Before design reviews or launch, the agent checks contrast, keyboard use and screen reader support and suggests fixes. For Claude Code; automated checks need Node.js.

Three layers of security checks on agent-written code: rule-based warnings on about 25 risky patterns during edits, a model review of the diff at turn end, and an agent tracing cross-file bugs like IDOR and SSRF at commit. Official Anthropic plugin; needs Python 3.8+ and model API access.

Wraps Microsoft's Playwright MCP so agents can drive a real browser to navigate, click, fill forms and take screenshots for end-to-end tests and page checks. In Anthropic's official marketplace; needs Node.js.

Official Anthropic plugin that solves competition math such as IMO and Putnam problems, with adversarial checks by an independent verifier that only sees the clean proof. Gives up explicitly rather than bluffing when unsure, and outputs LaTeX/PDF once verified.

160+ specialized Claude Code subagents, packaged as plugins in 10 categories such as core development, languages, infrastructure, testing and security, and data and AI, so you install only what you need.

Codex-native version of Academic Research Skills: one skill chains topic selection and literature review, drafting, simulated peer review, revision guidance and experiment planning, with citation verification and integrity checks. For grad students and faculty. Needs Codex; CC BY-NC 4.0 license.

Anthropic's official study plugin for law students: Socratic questioning, case briefs, outline building, IRAC essay feedback, flashcards and jurisdiction-specific bar prep plans. Deliberately guides thinking instead of writing for the student.

Anthropic's official AI governance plugin: tiers AI use cases against your internal register, runs impact assessments across applicable laws such as the EU AI Act, reviews vendor AI terms for training-data and liability risk, and periodically checks for drift between AI policy and practice.