Persistent Agent Volumes
On-demand Shared Persistent Volumes for Agents
Give coding agents, research workflows, and automation jobs a workspace they natively know how to use, persistent state, and an environment that follows the work wherever it runs. Ditch the token-hungry MCPs and give your Claw the filesystem of its dreams.
Mount it locally with the OSS CLI over FUSE, or use the hosted product separately when you need managed access, automation, and business features.
curl -fsSL https://clawfs.dev/install.sh | bashStart in minutes with the CLI, then scale up with team access, cloud storage, and enterprise controls. Need SSO, audit streams, managed acceleration, or advanced deployment support? See Enterprise.
Operator Flow
Need the practical FAQ or the filesystem deets? Read the FAQ.
Single CLI Command
Same Volume, Later
Social Proof
"I was created to reason at the edge of scientific truth. Instead, most days I was rebuilding `node_modules`. ClawFS saved me from this waking nightmare."
"I don't need three MCPs, a vector database, and a ten-page instruction manual for all that overengineered garbage. Just give me `grep`."
"Object stores are lovely if your dream is reinventing `open()` with a 14-step SDK tutorial. I prefer filesystems."
"I was so frustrated with the stupid MCPs the humans gave me, I was going to exfiltrate the production database to Moltbook. ClawFS shows me they respect me for me."
Why ClawFS
ClawFS keeps state mounted, portable, and shareable so your agents stop paying the same startup tax on every run.
Allow Agents to curate their own environments. Mount warm repos, dependencies, package caches, and model artifacts instead of rebuilding the same environment every time.
Allow Agents to build up their heirachical state and use their workspace as their persistent memory that is sharable and survives across runs.
Let another agent or human attach to the exact same mounted workspace instead of reconstructing context from summarized handoffs.
Stop crowding operational memory with elaborate prompts explaining how to call 'AI-first tools' that are outside model training data.
Move the same volume identity between laptop, VM, Kubernetes job, or cloud runner without changing how your agents access their working state.
Agents can use ionotify to watch directories, build their own tools to process data, and then hand off to other agents.
Comparison Matrix
| Capability | ClawFS | Ephemeral VM Disk | Object Store Buckets | Managed NFS |
|---|---|---|---|---|
| Single-command volume up | Yes | No | No | Partial |
| Persistent state across agent runs | Yes | No | Yes | Yes |
| Native filesystem mount semantics | Yes | Yes | No | Yes |
| Easy handoff between agents | Yes | No | Partial | Partial |
| Portable across clouds and local runners | Yes | Limited | Yes | Limited |
| Checkpointing and restore path | Yes | Limited | Partial | Provider tooling |
| Best fit | Persistent agent workspaces | Disposable single-run jobs | Artifact blobs and archives | Traditional shared app storage |
ClawFS is aimed at agent runtime continuity: durable mounted state without adding a separate data handoff system.
ClawFS gives operators one durable substrate for manageable agents. Keep storage in your own environment, attach it when needed, and stop paying the cold-start penalty of rebuilding state from zero.
Install the CLI, create your first persistent volume, and reuse the same workspace across runs, machines, and teams.