Stanford & Northeastern Built "Git for AI Agents" with Reversible Execution
rohanpaul_ai · x · 2026-08-10
Stanford and Northeastern developed Shepherd, a version control system designed specifically for AI agents, acting as a "Git for AI agents."
Existing agent frameworks typically only record transcripts and environment snapshots, meaning rewinds recover files but not live processes or provider caches. Shepherd solves this with core features:
- Commits an agent's running process and filesystem together.
- Makes execution revertible like Git: check out any past commit and resume from that exact state.
- Turns every model action, tool call, and environment change into a typed commit.
In multi-agent benchmarks, two coding agents sharing one repo typically score worse (28.8%) than a single agent doing both jobs alone (57.2%). By letting a supervisor watch runs live and rewind before a bad write lands, Shepherd closes 91% of that performance gap.
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