by Siddhant-K-code · MCP Server · ★ 54
agent-trace for AI agents. Capture and replay every tool call, prompt, and response from Claude Code, Cursor, or any MCP client. Why A coding agent rewrites 20 files in a background session. You get a pull request. You do not get the story. Which files did it read first? Why did it call the same tool three times? What failed before it found the fix? Most tools trace LLM calls. That is one layer. The gap is everything around it: tool calls, file operations, decision points, error recovery, the actual commands the agent ran. captures the full session and lets you replay it later.
| Stars | 54 |
| Forks | 5 |
| Language | Python |
| Category | MCP Server |
| License | MIT |
| Quality Score | 60.104/100 |
| Last Updated | 2026-05-20 |
| Created | 2026-03-15 |
| Platforms | claude-code, cli, gemini, mcp, python |
| Est. Tokens | ~42k |
Explore other popular mcp server tools:
agent-trace is strace for AI agents. Capture and replay every tool call, prompt, and response from Claude Code, Cursor, Gemini CLI or any MCP client. It is categorized as a MCP Server with 54 GitHub stars.
agent-trace is primarily written in Python. It covers topics such as agent-observability, agent-trace, agent-traces.
You can find installation instructions and usage details in the agent-trace GitHub repository at github.com/Siddhant-K-code/agent-trace. The project has 54 stars and 5 forks, indicating an active community.
agent-trace is released under the MIT license, making it free to use and modify according to the license terms.