Skip to main content

Toolchain

Instruction​

  • AGENTS.md: Open format for guiding coding agents.
  • llms.txt: Helping language models use website.
  • System: System prompts for agents.

SDK​

  • LangChain: Platform for reliable agents.
  • LangGraph: Build resilient language agents as graphs.
  • LlamaIndex: Build RAG applications on top of LLMs.
  • Crew: Orchestrate role-playing, autonomous agents
  • AgentOS: Build, run, manage multi-agent systems.
  • TypeChat: Makes it easy to build natural language interfaces.

RAG​

  • RAGFlow: Superior context layer for agents.

Agents​

  • Code: Coding agents.
  • Claw: Personal AI assistant.
  • CUA Infrastructure for computer-use agents.
  • SWE: Issue auto-fixer.
  • PR: Pull request reviewer.
  • Poke: NeurIPS 2025 Pokémon agent challenge.

Multi-Agent​

Memory​

  • Mem0: Universal memory layer.
  • MemOS: Memory operating system.
  • Beads: Git-backed graph issue tracker and persistent memory.

Workflow​

  • n8n: n8n workflows.
  • Dify: LLM app development platform combines workflow, RAG pipeline, agent capabilities.
  • GitHub: GitHub agentic workflows.

Project​

  • VibeKanban: Run coding agents in parallel without conflicts, and perform code review.

Frontend​

  • Vercel: Build AI-powered web applications.
  • Mastra: Build AI-powered applications and agents with modern TypeScript stack.
  • JSON: AI → JSON → UI.
  • Puck: Agentic visual editor for React.
  • Figma: Control Figma from command line.
  • Stitch: UI design toolkit.
  • Agentation: Elements visual feedback for agents.

Browser​

  • BrowserUse: Web agent.
  • Agent: Browser automation CLI.
  • Stagehand: Browser Automation Framework.
  • LightPanda: Headless browser designed for automation.
  • FireCrawl: Turn entire websites into LLM-ready markdown or structured data.
  • X-Crawl: Flexible Node.js AI-assisted crawler library.

Client​

Copilot​

Slide​

  • Banana: AI-native PPT generator based on nano banana pro.

Documentation​