I built a VS Code extension to visualize and debug LangGraph agents
A developer released the Agentic Engineering Extension, a free VS Code extension that turns LangGraph and CrewAI agent codebases into interactive graphs for visualization and debugging. It also runs real compiled graphs, supports visual editing that writes Python, and includes an evaluation engine plus Copilot-backed analysis commands.
- The free Agentic Engineering Extension renders LangGraph/CrewAI agent graphs in VS Code (also Cursor, Windsurf, Antigravity), with node-click source jumps, execution replay, and Python-writing visual edits.
- It includes an evaluation engine that runs a test dataset against the graph to flag regressions, plus 10 Copilot-backed commands citing file:line locations.
- Developer iroy2000 released the extension free on the VS Code Marketplace with a walkthrough at agentic-engineering-extension.vercel.app.
Related
xNode360: live infrastructure map where each node reports its own status
xNode360 is a live map of your infrastructure where every node self-reports status and latency. It now ships as a VS Code extension with a side-panel map and status-bar failure count, plus an MCP server so LLM assistants can query the live infra state. It targets developers whose setups have grown too complicated to track mentally.
xNode360 is a live map of your infrastructure where every node self-reports status and latency. It now ships as a VS Code extension with a side-panel map and status-bar failure count, plus an MCP server so LLM assistants can query the live infra state. It targets developers whose setups have grown too complicated to track mentally.
Building complex workflow with LangGraph
Medium tutorial walks through building a decision-making AI agent workflow with LangGraph
Medium tutorial walks through building a decision-making AI agent workflow with LangGraph
Skill Canvas - VS Code visual workflow editor for Claude Code agents and skills
A VS Code extension for sketching multi-step agent workflows on a canvas. Users place agents, skills, and tools, connect them with decision and retry paths, and export the diagram as standard Claude Code agent and skill files (also GitHub Copilot). It can import existing agents and skills for visual editing, solving the problem of managing complex agent configs by hand.
A VS Code extension for sketching multi-step agent workflows on a canvas. Users place agents, skills, and tools, connect them with decision and retry paths, and export the diagram as standard Claude Code agent and skill files (also GitHub Copilot). It can import existing agents and skills for visual editing, solving the problem of managing complex agent configs by hand.
Production-grade Agent Graph Engineering: a LangGraph project and blog series, distilled from 5 years
An engineer who has built realtime AI assistants since the ChatGPT beta published a working LangGraph-based agent architecture on GitHub plus a blog series explaining each design decision. Two parts are live with new installments every couple of days. The project targets practitioners who find existing LangGraph material stops at toy two-node examples.
An engineer who has built realtime AI assistants since the ChatGPT beta published a working LangGraph-based agent architecture on GitHub plus a blog series explaining each design decision. Two parts are live with new installments every couple of days. The project targets practitioners who find existing LangGraph material stops at toy two-node examples.
Samantha: open-source persistent state and memory layer for 24/7 LLM agents
A post-mortem of running autonomous LLM agents in production, identifying failure modes of in-memory agent state graphs like LangGraph and CrewAI. The authors open-sourced Samantha, which appears to address context window degradation, async task corruption, and state loss on crashes. It targets developers running autonomous agents continuously in production.
A post-mortem of running autonomous LLM agents in production, identifying failure modes of in-memory agent state graphs like LangGraph and CrewAI. The authors open-sourced Samantha, which appears to address context window degradation, async task corruption, and state loss on crashes. It targets developers running autonomous agents continuously in production.