agent-filesystem

What problem does it solve?

This Skill helps agents work with code and documents as if they were on a normal filesystem, while keeping workspace state stored remotely in Redis for easy checkpointing, restoring, and forking.

Core Features & Use Cases

• Mount workspaces for local work: Use AFS mounts so agents can run real tools against real directories while edits stay connected to the Redis-backed workspace.
• Support agent workflows via MCP and checkpoints: Interact with AFS through afs mcp and create durable restore points using explicit checkpoints (afs cp ...).
• Choose sync vs live mount: Use sync mode for a reconciled local directory, or mount mode for a live Redis-backed filesystem (NFS/FUSE depending on OS).

Use case: An agent needs to edit and run tests on a repository, search across prior notes, and later roll back to a known-good state—without tying the work to one machine’s local disk.

Quick Start

Ask the AI to create and mount a demo workspace so you can start editing files locally: make sure you have Redis available, then run: make commands, ./afs auth login, ./afs ws create demo, ./afs ws mount demo ~/demo, and cd into ~/demo.