
See why your coding agent looped, drifted, or burned tokens
agentfdr reads the transcripts Claude Code already writes and turns them into a flight recording: per-turn timeline, automatic anomaly flags (loops, error streaks, token burns), live watch mode, plan-usage tracking, and cost estimates. Zero instrumentation, 100% local, MIT.
agentfdr is an open-source developer tool that analyzes coding transcripts to create a detailed timeline of agent performance, highlighting anomalies such as loops and token burns. It operates locally without instrumentation and is licensed under the MIT license.