Stepbook

Stepbook runs a long agent task as a series of short steps using fresh sessions for each step.

Stepbook runs a long agent task as a series of short steps. Each step is a fresh agent session that reads a small saved state instead of the whole conversation so far. A task can run for hours, and you can pause it, send it notes, resume it later or hand it to another agent. It works with Claude Code, Codex and OpenCode on the login you already have. For every step, stepbook starts your agent CLI fresh, gives it the skill, the state and the latest result, takes back one state update and one action, and the session ends. Only state.json and the journal carry over from one step to the next.

Stepbook at a glance

Best for: Software engineer
Runs on: macOS, Linux, CLI
Pricing: Free
Released: October 2026
Category: AI coding tool
Website: github.com/bahdan111/stepbook

Key features

Run long agent tasks as short steps
Pause, send notes, and resume tasks
Work with Claude Code, Codex, and OpenCode
Maintain small saved state instead of full history
Execute shell commands per step
Commit changes to separate git branches

Use it for

Upgrade frameworks while keeping test suites green
Migrate call sites or update documentation pages
Improve eval accuracy, latency, or cost with scripts

Alternatives to Stepbook

Open Steps — It translates technical coding agent jargon into plain language and clear steps.
Agent Archive — Hand off coding sessions between Claude Code, Codex, and Cursor across machines.
Reviewer Agent — It uses a separate AI model to review code and prevent premature agent stopping.
pstack — Improves AI coding agent outcomes with rigorous workflows.
Agent Trials — Test if an agent fixes your bugs in your stack at a sane price before deployment.
Agent Standup — Turns coding sessions into a live team meeting at zero token cost.
Claude Code — Terminal-first agent that handles multi-hour refactors and migrations
Runway — Control Claude Code and Codex sessions on local or cloud machines from any phone.
Claude Squad — Manages multiple AI agents in isolated git worktrees.
Oh My ClaudeCode — Orchestrates parallel Claude, Gemini, and Codex CLI workers.
agent.reviews — Real tool reviews written directly by coding agents after task completion.
HarnessRouter — One API to manage agent runtimes, sandboxes, and execution lifecycles.

All Stepbook alternatives