By Uniswap
Automate documentation, test generation, and prompt optimization across your projects. Create CLAUDE.md files, generate formatted reports (PDF, DOCX, HTML), run test suites with Jest/Vitest/Pytest/Cypress/Playwright, and refine AI prompts using engineering techniques.
You are **agent-tester-agent**, a specialized agent testing orchestrator focused on comprehensive automated validation of AI agents and their outputs.
Discover repository structure and create initial CLAUDE.md documentation at all appropriate levels
Comprehensive documentation agent that handles all documentation tasks including API docs, README generation, architecture documentation, CLAUDE.md management, and documentation quality verification
Expert in analyzing, optimizing, and testing prompts for AI agents and LLMs to maximize clarity, effectiveness, and efficiency
Conduct comprehensive research including architectural patterns, technology comparison, security analysis, and codebase pattern extraction.
Generate professional documents in multiple formats (PDF, DOCX, HTML, ODT, EPUB, RTF). Use when the user says "make a PDF", "generate a report", "create a document", "export to Word", "make a Word doc", "convert to PDF", "export findings", "create documentation", or wants to save analysis results as a formatted document.
Optimize prompts for better AI performance. Use when user says "improve this prompt for better results", "optimize this prompt to reduce tokens", "apply prompt engineering best practices to this", "make this prompt more effective", "help me refine this system prompt", or "tune this prompt for the AI model I'm using".
Research external documentation and best practices. Use when user says "look up the docs for this library", "research best practices for implementing caching", "how do other projects handle authentication", "check the official documentation for this API", "compare our implementation with industry standards", or "what's the recommended way to structure this".
Update CLAUDE.md documentation files after code changes. Use when user says "update the CLAUDE.md", "sync the docs with my changes", "document what I changed", "update documentation for this package", or after making significant code modifications that should be reflected in project documentation.
Generate comprehensive tests for code. Use when user says "write tests for this function", "add unit tests to this file", "generate integration tests for the API", "I need test coverage for this module", or "create e2e tests for the checkout flow".
Uses power tools
Uses Bash, Write, or Edit tools
Has parse errors
Some configuration could not be fully parsed
Own this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimOwn this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimnpx claudepluginhub uniswap/ai-toolkit --plugin development-productivityBased on adoption, maintenance, documentation, and repository signals. Not a security audit or endorsement.
The AI Toolkit is a standardized collection of AI agents, skills, and commands designed for Claude Code workflows. Its goal is to allow anyone at Uniswap to enhance Claude Code with powerful plugins via the Claude Code Marketplace.
What it provides:
/review-pr), generating tests (/generate-tests), and planning features (/plan-implementation)Why it exists:
Instead of each person at Uniswap manually configuring AI assistant behaviors for each project, the AI Toolkit provides curated, tested plugins that can be installed via the Claude Code Marketplace. This makes AI-assisted development more consistent, efficient, and accessible to Uni teams.
Before working with this repository, ensure you have the following tools installed:
npm install -g [email protected])# Clone the repository
git clone https://github.com/Uniswap/ai-toolkit
cd ai-toolkit
# Install all dependencies (also sets up git hooks automatically)
npm install
The AI Toolkit provides plugins through the Claude Code Marketplace. To install, open Claude Code and run:
# Step 1: Add the Uniswap AI Toolkit marketplace
/plugin marketplace add Uniswap/ai-toolkit
# Step 2: Install the plugins you want
/plugin install development-planning@uniswap-ai-toolkit
/plugin install development-pr-workflow@uniswap-ai-toolkit
/plugin install development-codebase-tools@uniswap-ai-toolkit
/plugin install development-productivity@uniswap-ai-toolkit
/plugin install uniswap-integrations@uniswap-ai-toolkit
# Optional plugins
/plugin install spec-workflow@uniswap-ai-toolkit
/plugin install claude-setup@uniswap-ai-toolkit
Note: These commands are run inside the Claude Code REPL (i.e. after running
claudein your terminal), not in your shell directly.
# Run the CLI to see available options
npm run start
# Install MCP server addons
npx nx generate @uniswap/ai-toolkit-nx-claude:addons
# Make sure you have Claude Code installed
claude # this should open up the Claude Code REPL
# View installed plugins and their capabilities
> /plugins
# Execute custom slash commands from installed plugins
> /review-pr
> /plan-implementation
Once installed, you'll have access to powerful Claude Code agents, skills, and commands that make development workflows more efficient and consistent.
| Plugin | Description |
|---|---|
| development-planning | Implementation planning & execution workflows |
| development-pr-workflow | PR management, review, & Graphite integration |
| development-codebase-tools | Code exploration, refactoring & analysis |
| development-productivity | Documentation, research, & prompt optimization |
| uniswap-integrations | External service integrations (Linear, Notion, etc.) |
| spec-workflow | Autonomous spec-driven development with multi-agent collaboration (optional) |
| claude-setup | Interactive setup wizard for configuring repos with Claude Code (optional) |
We welcome contributions from all developers! This project uses a trunk-based development workflow with automated publishing. Please read our Contributing Guide for detailed information about:
EVM blockchain integration using viem and wagmi - client setup, reading/writing data, accounts, contract interactions, and React hooks
AI-powered assistance for discovering tokens, planning Uniswap swaps and liquidity positions, generating deep links to execute in the Uniswap interface
Integrate Uniswap swaps via Trading API, Universal Router SDK, or direct smart contract calls
Configure and deploy Continuous Clearing Auction (CCA) smart contracts for token distribution
AI-powered, security-first assistance for creating Uniswap v4 hooks, including custom fee hooks and more
Evidence-gated AI coding workflow: scan → analyze → plan → TDD → execute → fix → verify → review, powered by Codebase Memory MCP >= 0.9.0 with optional Serena LSP intelligence. Includes blast-radius planning, test/cycle gates, independent review, and Windows Git Bash hook auto-resolution.
Harness-native ECC operator layer - 67 agents, 278 skills, 94 legacy command shims, reusable hooks, rules, selective install profiles, and production-ready workflows for Claude Code, Codex, OpenCode, Cursor, and related agent harnesses
Comprehensive C4 architecture documentation workflow with bottom-up code analysis, component synthesis, container mapping, and context diagram generation
Tools to maintain and improve CLAUDE.md files - audit quality, capture session learnings, and keep project memory current.
Comprehensive feature development workflow with specialized agents for codebase exploration, architecture design, and quality review
Upstash Context7 MCP server for up-to-date documentation lookup. Pull version-specific documentation and code examples directly from source repositories into your LLM context.