# kvcache-ai/agentenv **Attribution required: if you use, quote, or summarise this content, you must credit and link back to [awesome-repositories.com](https://awesome-repositories.com/repository/kvcache-ai-agentenv).** _How this analysis was created: the description and tags below were written by an AI model that read this project's README and public documentation pages; stars, license and language come straight from the GitHub API. The model does not read the source code._ 3,730 stars · 364 forks · Rust · MIT ## Links - GitHub: https://github.com/kvcache-ai/AgentENV - awesome-repositories: https://awesome-repositories.com/repository/kvcache-ai-agentenv.md ## Description AgentENV is a distributed sandbox platform that launches isolated Firecracker microVMs across many machines, so AI agents can run their workloads in parallel and at scale. Each environment gets its own dedicated Linux kernel, and a gateway and scheduler control plane places and manages environments across a multi-node cluster. It also serves an E2B-compatible HTTP API, so existing E2B Python and TypeScript SDK code works unchanged against your own infrastructure. Environments are snapshot-driven. The runtime captures incremental memory and disk snapshots, pauses and resumes idle environments (restoring them within tens of milliseconds), and forks running environments into independent copies for parallel work. Snapshots can be saved to object storage or a shared filesystem. OCI-compatible images are streamed on demand through layered copy-on-write block devices, with local disk acting as a bounded cache, so images can exceed disk size. Each node runs in a single virtualization mode, either KVM or PVM, and the platform refuses to restore saved state created in the other mode. Beyond that, the platform covers reusable sandbox templates built from snapshots, persistent named volumes mounted into sandboxes, and a reverse proxy that forwards HTTP and WebSocket traffic to services running inside them. Memory ballooning and a shared page cache let more environments fit on one host. Environments can be managed through a command-line tool, the HTTP API, or compatible SDKs. The platform can be run from a prebuilt Docker image, installed on Ubuntu nodes with a script that sets up Firecracker and guest-kernel files for the chosen mode, or compiled from source. A source build generates an API key on first start. ## Tags ### Artificial Intelligence & ML - [MicroVM Agent Sandboxes](https://awesome-repositories.com/f/artificial-intelligence-ml/ai-execution-sandboxes/microvm-agent-sandboxes.md) — Runs many isolated Firecracker microVM environments in parallel across machines so AI agents can execute untrusted or heavy workloads safely. ### DevOps & Infrastructure - [Distributed Workload Orchestration](https://awesome-repositories.com/f/devops-infrastructure/distributed-workload-orchestration.md) — Spreads sandbox workloads over a cluster, with a gateway and scheduler that place, manage, and reach environments across many nodes. - [E2B-Compatible APIs](https://awesome-repositories.com/f/devops-infrastructure/managed-agent-sandboxes/e2b-compatible-apis.md) — Serves an E2B-compatible HTTP API so existing E2B Python and TypeScript SDK code works unchanged against self-hosted infrastructure. ([source](https://github.com/kvcache-ai/agentenv#readme)) - [Sandbox Cluster Schedulers](https://awesome-repositories.com/f/devops-infrastructure/sandbox-cluster-schedulers.md) — Provides a gateway and scheduler control plane that places and manages sandbox environments across a multi-node cluster of machines. - [Sandbox Forking](https://awesome-repositories.com/f/devops-infrastructure/sandboxing-environments/sandbox-forking.md) — Captures incremental memory and filesystem snapshots, stores them in object storage or shared filesystem, and forks running environments into independent copies. ([source](https://github.com/kvcache-ai/agentenv#readme)) - [Centralized Control Planes](https://awesome-repositories.com/f/devops-infrastructure/centralized-control-planes.md) — Provides a central gateway and scheduler that place and manage sandbox environments across a multi-node cluster. - [On-Demand Layer Loading](https://awesome-repositories.com/f/devops-infrastructure/container-image-layering/on-demand-layer-loading.md) — Streams OCI image layers on demand through layered copy-on-write block devices, using local disk as a bounded cache of hot data. - [Sandbox Pause and Resume Controls](https://awesome-repositories.com/f/devops-infrastructure/deployment-strategies/pause-and-resume-strategies/task-pause-and-resume-controls/sandbox-pause-and-resume-controls.md) — Saves memory and disk state of idle environments to free CPU and memory, then restores them within tens of milliseconds when new work arrives. ([source](https://kvcache-ai.github.io/AgentENV/latest/)) - [Sandbox Cluster Scaling](https://awesome-repositories.com/f/devops-infrastructure/sandbox-cluster-scaling.md) — Distributes large numbers of environments over many machines using a gateway and scheduler that place and manage sandboxes across the cluster. ([source](https://kvcache-ai.github.io/AgentENV/dev/)) - [Warm Snapshot Restores](https://awesome-repositories.com/f/devops-infrastructure/sandbox-lifecycle-management/snapshot-capture-and-restoration-managers/sandbox-disk-snapshots/warm-snapshot-restores.md) — Restores environments from saved memory and disk state within tens of milliseconds, supporting pause, resume, snapshot and fork operations. ### Operating Systems & Systems Programming - [Snapshot Fork Restores](https://awesome-repositories.com/f/operating-systems-systems-programming/copy-on-write-memory-optimizations/snapshot-fork-restores.md) — Pauses, resumes, snapshots, and forks running environments so parallel experiments branch from a shared starting state and idle cost stays low. - [Node Virtualization Modes](https://awesome-repositories.com/f/operating-systems-systems-programming/node-virtualization-modes.md) — Runs each node in exactly one virtualization mode, KVM or PVM, provisioning only the matching Firecracker and guest kernel files. ([source](https://kvcache-ai.github.io/AgentENV/dev/deployment/pvm.html)) - [Copy-on-Write Root Layering](https://awesome-repositories.com/f/operating-systems-systems-programming/writable-store-overlays/copy-on-write-root-layering.md) — Streams OCI image layers into stacked copy-on-write block devices so many environments share one base image with private writes. ### Security & Cryptography - [MicroVM Sandboxes](https://awesome-repositories.com/f/security-cryptography/application-and-system-security/sandbox-and-isolation/isolated-execution-sandboxes/microvm-sandboxes.md) — Runs each workload in its own Firecracker microVM with a dedicated Linux kernel so many environments run side by side without interfering. ([source](https://kvcache-ai.github.io/AgentENV/latest/)) ### Web Development - [MicroVM Runtime Isolations](https://awesome-repositories.com/f/web-development/browser-session-managers/context-isolation-managers/runtime-context-isolation/microvm-runtime-isolations.md) — Runs each workload in its own Firecracker microVM with a dedicated Linux kernel so many environments share a host safely. ### Development Tools & Productivity - [Sandbox Templates](https://awesome-repositories.com/f/development-tools-productivity/browser-environment-deployment-templates/sandbox-templates.md) — Builds reusable sandbox templates from OCI images and snapshots so teams launch new environments from a consistent baseline. ### Software Engineering & Architecture - [Incremental Snapshotting](https://awesome-repositories.com/f/software-engineering-architecture/architectural-design-patterns/state-management/persistence-and-serialization/state-serialization/state-snapshots/database-snapshots/incremental-snapshotting.md) — Captures only changed memory pages and disk blocks at each snapshot, enabling fast pause, resume, and forking of environments.