WinnCook/agentcasebook
Local-first session review for Claude Code and Codex.
Platform-specific configuration:
{
"mcpServers": {
"agentcasebook": {
"command": "npx",
"args": [
"-y",
"agentcasebook"
]
}
}
}Add the config above to .claude/settings.json under the mcpServers key.
[](https://badge.fury.io/js/agentrace) [](https://github.com/WinnCook/agentrace/actions) [](https://opensource.org/licenses/MIT) [](https://nodejs.org/)
Universal agent session tracing, replay, and analysis. Think "OpenTelemetry for AI agents" but lightweight and opinionated.
npx agentrace list --scan # Discover and list all agent sessions
npx agentrace stats --last 7d # Show analytics for last week
npx agentrace replay <session> # Step-by-step session replay
npx agentrace export <session> --format markdown # Export session report*Works out of the box with Claude Code, Codex, OpenClaw, and more.*
šÆ Zero config ā Auto-detects agent sessions from known paths š Universal tracing ā Supports Claude Code, Codex, OpenClaw, Cursor š Rich analytics ā Cost tracking, pattern detection, success rates š¬ Session replay ā Step-by-step visualization with file changes š° Cost analysis ā Token usage and spend tracking by model/agent š Pattern detection ā Identifies loops, failures, and inefficiencies š File tracking ā Monitor code changes across sessions š MCP server ā Query your session history programmatically š Export formats ā Markdown, JSON, HTML reports šļø Local-first ā SQLite storage, no cloud dependencies
npx agentrace --helpnpm install -g agentrace
agentrace --helpbrew install agentraceagentrace list [options]List and discover agent sessions.
agentrace list # Last 7 days
agentrace list --last 3Loading reviews...