Run an AI coding agent and review its proposed changes before applying them
Grant an agent read-only or read-write access to specific repositories only
Replay or fork a past agent run from its recorded execution trace
Try the framework offline with a deterministic provider, no API key needed
| shepherd-agents/shepherd | amap-cvlab/abot-world | aminblg/simpleenglish | |
|---|---|---|---|
| Stars | 2,290 | 2,295 | 2,280 |
| Language | Python | Python | Python |
| Setup difficulty | moderate | hard | easy |
| Complexity | 4/5 | 5/5 | 1/5 |
| Audience | developer | researcher | developer |
Figures from each repo's GitHub metadata at analysis time.
Live agent runs need the Claude CLI with a subscription or API key, an offline mode needs neither.
Shepherd is a framework for running AI agents in a way that can be inspected, undone, and reviewed before anything actually changes your files. When an agent completes a task, its work is not written directly into your project. Instead it comes back as a proposal that sits to one side, which you can look at, run, and only apply to your real files once you have decided you actually want it. The core idea is that a task is written as a plain Python function with no body at all. Its name, its parameters, and its docstring describe what the agent should do, and those parameters also define exactly what the agent is allowed to touch. For example, marking a parameter as a writable repository handle grants the agent permission to write to that specific location, while marking it read only means any attempt by the agent to write there is blocked at the operating system level, on both macOS and Linux. This means you can read a task's function signature and know precisely what it is and is not permitted to do, before it ever runs. Every agent run is recorded as a durable trace, so its outputs can be selected and merged in, applied on top of a workspace that has since changed, or discarded entirely, with the record kept either way. Shepherd includes an offline quickstart that needs no API key and runs the same underlying machinery using a deterministic built in provider instead of a real AI model, useful for trying the system out. There is also a live quickstart that uses the Claude command line tool, which requires either a Claude subscription or an Anthropic API key. The project describes itself as being built specifically so that meta-agents, meaning higher level agents, can supervise, optimize, and train other agents by observing, forking, and replaying their execution. It is installed with pip, requires Python 3.11 or newer, and the README states plainly that it is in early alpha with APIs that may still change between releases. Windows is not supported directly, and the documentation recommends using WSL instead.
A framework that runs AI agents as reviewable, reversible proposals, with function signatures defining exactly what each agent can read or write.
Mainly Python. The stack also includes Python, Claude CLI, Landlock.
Setup difficulty is rated moderate, with roughly 30min to a first successful run.
Mainly developer.
This repo across BitVibe Labs
Don't trust strangers blindly. Verify against the repo.