Skip to content
Get packs

Manage Mounts

Control which folders on your computer NanoClaw agent containers can read or write.

Source Download skill
Repository
nanocoai/nanoclaw
Stars
31,072
Licence
MIT
Files
1 file

What it does

NanoClaw runs AI agents inside isolated containers, and by default those containers cannot see your computer's files. This skill lets the agent view, add, or remove entries in a mount allowlist, the list of host folders containers are allowed to access. For each folder it asks whether access should be read-only (the safer default) or read-write, then saves the config through NanoClaw's setup script.

Great for

  • Developers running NanoClaw who want agents to work on a specific project folder
  • Anyone who needs to audit or tighten which host directories agents can reach
  • Users who want to reset container file access back to nothing

Skip it if

  • Users who do not run NanoClaw, since the skill only configures NanoClaw containers (it would be irrelevant elsewhere)
  • Granting read-write access to sensitive folders such as home directories, SSH keys, or credential stores, because agents could modify or leak those files

Try asking

  • “Show me which directories my NanoClaw agents can currently access”
  • “Give the agent read-only access to ~/projects/website”
  • “Add read-write access to ~/work/app-repo for my agent containers”
  • “Remove the /tmp/scratch entry from the mount allowlist”
Step-by-step guide: how to use Manage MountsInstall, first run, examples, tips and FAQs

Install in 3 steps

  1. 1

    Download the skill

    The ZIP contains the 1 original files, the MIT licence and an attribution note.

    manage-mounts.zip
  2. 2

    Add it to your agent

    Claude Code (all projects)

    mkdir -p ~/.claude/skills && unzip manage-mounts.zip -d ~/.claude/skills/

    Claude Code / Agent SDK (this project only)

    mkdir -p .claude/skills && unzip manage-mounts.zip -d .claude/skills/

    Claude apps (web/desktop): Settings → Capabilities → Skills → Upload skill, and choose the ZIP.

  3. 3

    Ask for the task

    No special command needed — the agent reads the skill's description and loads it when your request matches. Try one of the prompts above.

Get better results

  • Prefer read-only access unless the agent truly needs to write files, and review the list after each change.
  • Avoid mounting broad folders like your whole home directory; mount only the specific project folder needed.
  • After changing mounts, restart the affected group's container with ncl groups restart if it is already running, since new containers pick up changes automatically.

What's inside

  • SKILL.md

Redistributed unmodified from nanocoai/nanoclaw under the MIT licence. The assessment was drafted with AI from the skill's files on 2026-10-29. Skills can instruct your agent to run code — read SKILL.md and any scripts before installing.