AgenticLens turns agent logs into DevTools
Agent debugging today is mostly scrolling through console logs trying to figure out what the hell happened. AgenticLens fixes that. It launched today as a local, open-source visual debugger for the Claude Agent SDK — flow graphs, timelines, execution replay, all from your existing log files.
Built by Amit Mishra, the pitch is simple: DevTools for agents. You see where latency came from, which tool call misfired, and you can replay the run instead of re-running it. Everything stays on your machine, no logs uploaded.
Why this matters: every team running agents in production has written some janky Python script to grep logs. As agent orchestration gets more complex — subagents, parallel tool calls, skills — that approach falls apart. AgenticLens is the browser DevTools moment for agent engineering. First-mover advantage on an obvious primitive.
Link: https://agenticlens.in/
← Back to all articles
Built by Amit Mishra, the pitch is simple: DevTools for agents. You see where latency came from, which tool call misfired, and you can replay the run instead of re-running it. Everything stays on your machine, no logs uploaded.
Why this matters: every team running agents in production has written some janky Python script to grep logs. As agent orchestration gets more complex — subagents, parallel tool calls, skills — that approach falls apart. AgenticLens is the browser DevTools moment for agent engineering. First-mover advantage on an obvious primitive.
Link: https://agenticlens.in/
Comments