Record and replay AI agent execution for debugging
Record and replay AI agent execution for debugging. Exposed over MCP by the agent mcp server, that capability becomes something an assistant can invoke while it works, not something you go and do afterwards.
MCP server for agent session recording and replay — debug non-deterministic agent behavior with session comparison and divergence detection.
agent-replay-mcp on npm is all you need. Most clients run it directly, so configuration is a few lines and a restart.
Everything the assistant can do here goes through one of these:
log_action — capture input, output, reasoning, timingreplay_session — review what happened step by stepcompare_sessions — diff today's run vs yesterday'sfind_divergence_point — pinpoint where it went wrongexport_session — JSON for tooling, markdown for humansrecord_session — The record_session tool exposed by this serverstop_recording — Returns: action count, total duration, action type breakdownPlenty of developer tooling servers cover similar ground. The differences that matter in practice are scope of access and how much setup stands between you and a working tool call. Agent's toolset — log_action, replay_session, compare_sessions and 4 more — is a fair guide to whether it matches your workflow. It is maintained by mdfifty50-boop; worth a glance at recent repository activity before you build anything load-bearing on it.
SyncDev reviews every entry in this directory against the project's own documentation before publishing, and revisits them as servers change.
| Tool | What it does |
|---|---|
| log_action | capture input, output, reasoning, timing |
| replay_session | review what happened step by step |
| compare_sessions | diff today's run vs yesterday's |
| find_divergence_point | pinpoint where it went wrong |
| export_session | JSON for tooling, markdown for humans |
| record_session | The record_session tool exposed by this server. |
| stop_recording | Returns: action count, total duration, action type breakdown. |
{
"mcpServers": {
"agent-replay": {
"command": "npx",
"args": ["-y", "agent-replay-mcp"]
}
}
}Add to claude_desktop_config.json, then restart Claude Desktop.
Kill hallucinated APIs — version-accurate, up-to-date library documentation injected straight into context.
Microsoft's official browser automation server — drive a real browser through the accessibility tree, no screenshots needed.
GitHub's official server — repos, issues, pull requests, Actions and code security, straight from your assistant.
Issue tracking at the speed of conversation — Linear's official hosted server with OAuth and zero install.
Local repository surgery — status, diffs, commits, branches and history for any repo on disk.
Timezone sanity for AI — current time anywhere and correct conversions, without the model doing date math.