Stanford & Northeastern Unveil Shepherd: Git-like Version Control for AI Agents
rohanpaul_ai · x · 2026-08-10
Stanford and Northeastern have developed Shepherd, an open-source project acting as "Git for AI agents."
Existing agent frameworks typically only record transcripts and environment snapshots, meaning rewinding recovers files but not the live process or provider cache. Shepherd solves this by:
- Committing an agent's running process and filesystem together, making execution fully revertible.
- Using a copy-on-write fork mechanism that is 5x faster than docker commit.
- Achieving 95% KV-cache reuse during replays.
The runtime substrate is built for meta-agents, enabling them to observe, fork, replay, and revert runs to supervise, optimize, and train other agents.
Related event: Stanford Open-Sources Shepherd: Git for AI Agents(3 posts)→
More from coding & agent
- Opus 5.5 + Three.js demo stuns with water physics, fish and blossoms — nikola_mr64990 · 2026-10-16
- Karpathy's knowledge engine workflow: Claude Code turns raw notes into a self-updating wiki — nikola_mr64990 · 2026-10-16
- Sierra and Meta unveil Personal Agent Protocol with Zapier as launch partner — sujingshen · 2026-10-16
- ArcKit adds 102 AI-narrated short films for commands across 14 community overlays — mcraddock · 2026-10-16
- Personal agents live or die on chaining tasks, not single-task ability — sujingshen · 2026-10-16
- Dev laments durable coding agent crumbling on restarts with long subagent sessions — lucasmeijer · 2026-10-16