Tracely
LLM observability and CI/CD for AI agents with automatic regression testing
Visit tracely-ai.com ↗What is Tracely?
Tracely is a trace-native CI/CD platform for AI agents that automatically converts production failures into regression tests. It captures every agent trace via OTLP, grades runs with LLM-as-judge evaluators, clusters failures into issues, and freezes bad runs into hermetic test cases that block pull requests. The platform supports any agent stack including OpenAI, Anthropic, LangChain, and LangGraph.
Designed for teams building production AI agents, Tracely eliminates hand-authored test datasets by deriving all quality signals from real production traces. Failed runs are replayed in CI with recorded tool and model fixtures at zero cost, and the system alerts immediately when regressions occur. The project is built with Claude and available as self-hosted or managed SaaS.
Momentum
Daily activity over the last 30 days: upvotes, unique visits, new GitHub stars and Hacker News points.
How do we know it's vibe coded?
vibetrend discovered Tracely automatically. Evidence: GitHub Jwuthri/Tracely-ai; repo has .claude, CLAUDE.md. View source.
Similar vibe-coded apps
#1 this weekOpen-source local voice cloning, dubbing, transcription, and audiobook creation
#2 this week
#5 this weekSlash commands and agents for engineers using AI without compromising standards
#6 this week523 free lessons building AI systems from math and first principles
#8 this week
#10 this weekAI-powered design intelligence toolkit for building professional interfaces

Manage teams of AI agents to run your business from one dashboard

Persistent memory for AI agents across Claude Code, Cursor, and MCP clients