Fork, Explore, Commit: OS Primitives for Agentic Exploration
Not on the map. Why?
Abstract (arXiv)
AI agents increasingly perform agentic exploration: pursuing multiple solution paths in parallel and committing only the successful one. Because each exploration path may modify files and spawn processes, agents require isolated environments with atomic commit and rollback semantics for both filesystem state and process state. We introduce the branch context, a new OS abstraction that provides: (1) copy-on-write state isolation with independent filesystem views and process groups, (2) a structured lifecycle of fork, explore, and commit/abort, (3) first-commit-wins resolution that automatically invalidates sibling branches, and (4) nestable contexts for hierarchical exploration. We realize branch contexts in Linux through two complementary components. First, BranchFS is a FUSE-based filesystem that gives each branch context an isolated copy-on-write workspace, with O(1) creation, atomic commit to the parent, and automatic sibling invalidation, all without root privileges. BranchFS is open sourced in https://github.com/multikernel/branchfs, along with a Python integration library, BranchContext, that provides ready-to-use agent exploration patterns. Second, branch() is a proposed Linux syscall that spawns processes into branch contexts with reliable termination, kernel-enforced sibling isolation, and first-commit-wins coordination. Preliminary evaluation of BranchFS shows sub-350 us branch creation independent of base filesystem size, and modification-proportional commit overhead (under 1 ms for small changes).
Key facts. Cited by 5 papers in the map (data through 13 Oct 2026). First posted on arXiv 9 Feb 2026.
Builds on
No reference list yet.
Built on by
Citing papers per year: 2026: 5.
5 papers.
- 2026Shepherd: Enabling Programmable Meta-Agents via Reversible Agentic Execution Tracescited by 11
- 2026ACRFence: Preventing Semantic Rollback Attacks in Agent Checkpoint-Restorecited by 10
- 2026Safe to Resume? Breaking Execution Continuity of Agent Execution via Rollbackcited by 2
- 2026Sandlock: Confining AI Agent Code with Unprivileged Linux Primitivescited by 1
- 2026AgileLog: A Forkable Shared Log for Agents on Data Streamscited by 0
- 2026Safe to Resume? Breaking Execution Continuity of Agent Execution via Rollback29 Aug 2026
- 2026Sandlock: Confining AI Agent Code with Unprivileged Linux Primitives25 May 2026
- 2026Shepherd: Enabling Programmable Meta-Agents via Reversible Agentic Execution Traces11 May 2026
- 2026AgileLog: A Forkable Shared Log for Agents on Data Streams16 Apr 2026
- 2026ACRFence: Preventing Semantic Rollback Attacks in Agent Checkpoint-Restore21 Mar 2026