whereismyai
AgentTM
AgentTM is a time-travel debugger for AI coding agents built in Go that records and inspects noninteractive runs.
AgentTM is a local Go command-line tool that records real, managed noninteractive Codex runs in SQLite. It includes baseline and final workspace checkpoints, historical file inspection, and an embedded React and TypeScript user interface. Developers use it to review stored sessions, timelines, historical files, parent diffs, and isolated worktree forks. The embedded interface features a read-only Monaco viewer to inspect changed paths and unified patches. It requires Git and an authenticated Codex CLI to record sessions.
AgentTM at a glance
- Best for: Software engineer
- Runs on: CLI, Web, Windows, macOS, Linux, self-hosted, open source
- Pricing: Free
- Released: November 2026
- Category: AI coding tool
- Website: github.com/jyotishmoy12/agenttm
Key features
- Record noninteractive Codex runs in SQLite
- Inspect historical files and parent diffs
- Serve embedded React and TypeScript user interface
- Create isolated worktree forks
- Manually observe current workspace checkpoints
- List sessions, timestamps, status, and exit codes
- Redact known credential patterns
Use it for
- Debug AI coding agent sessions
- Inspect historical file states and checkpoints
- Compare parent diffs and unified patches
- Restore isolated worktree forks
Alternatives to AgentTM
- AGENTS.md — It separates agent-specific instructions from human-focused README files.
- Agent Manager — Run multiple AI coding agents side by side with live status, quick prompts, and diff review.
- Agentty — A single 24.4 MB static binary with native speed and no runtime dependencies.
- gptme — It works in your local shell without a sandbox, providing direct file access.
- Agent Archive — Hand off coding sessions between Claude Code, Codex, and Cursor across machines.
- AgentMemory — It runs locally with zero external databases and achieves high retrieval recall.
- AgentsView — It keeps your AI coding session archive local by default while tracking costs and activity.
- Agent Beacon — Retains coding agent knowledge across tools so errors do not repeat.
- Agent OS — Manage AI coding sessions side by side and check on them from your phone.
- Tigerless Agent Memory — Maintains agent history in portable, greppable Markdown files without API keys.
- Mdt — Avoids unnecessary rereading, token use, and distraction for agents.
- AgentOps — Track token usage, costs, and traces across 400 plus LLMs and agent frameworks.
All AgentTM alternatives