Skip to content
Back to skills

Workspace

ASecurity

Make a new workspace or check one in the Portfolio: plan the create-only install, apply it on a yes, and run the read-only structure check. Use for /workspace new, /workspace check, or when someone says set up a workspace, add a workspace, or is this workspace healthy.

  • 7 stars
  • 0 votes
  • 0 copies
  • 0 views
  • Added November 7, 2026
ai-agentspythongo

Security analysis

A100/100

Pro scans all 2 files and shows the line behind each finding

Scanned November 7, 2026

npx -y skills add foxrun-ai/foxrun --skill workspace --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Workspace?

Add the live security badge to your README. It updates with every re-scan.

Security grade badge for Workspace
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/foxrun-ai-workspace/badge)](https://www.skillsdirectory.com/skills/foxrun-ai-workspace)

More formats (shields.io, HTML) on the badges page. Keep it an A: scan every change in CI with Pro.

Download with Pro
SKILL.md
---
type: note
workspace: system
timestamp: 2026-10-26T14:30:00-04:00
name: workspace
description: "Make a new workspace or check one in the Portfolio: plan the create-only install, apply it on a yes, and run the read-only structure check. Use for /workspace new, /workspace check, or when someone says set up a workspace, add a workspace, or is this workspace healthy."
kind: knowledge
argument-hint: "new <name> [personal | ecom] | check <name>"
---

# workspace

A workspace is one folder in the Portfolio, `~/foxrun/<ws>/`, with its own `AGENTS.md`.
`new` plans first and writes only on the person's yes, because a new workspace is hard to take back.
`check` only reads. It never edits the workspace.

## Steps

1. **New: check the name.** Use lowercase kebab case. Run `ls ~/foxrun/` and `cat ~/foxrun/AGENTS.md`. If the name is taken, or another workspace already owns the subject, stop and say so. Done when the name is free and no workspace owns the subject.
2. **New: plan.** Run `python3 ~/.foxrun/base/install_workspace.py --target-workspace <ws> --workspace-type <personal|ecom> --by <you> --plan-out <scratch>/workspace-plan.json`. Run it with `--help` for the module options. Show the person the folders and files the plan makes. Done when the person said yes to that plan.
3. **New: apply.** Run `python3 ~/.foxrun/base/install_workspace.py --apply <scratch>/workspace-plan.json`. Done when `~/foxrun/<ws>/AGENTS.md` exists.
4. **Check.** Run `python3 ~/.foxrun/base/check_workspace.py <ws>` for JSON, or add `--output-prefix <outside folder>/<name>` to keep a report. Exit 0 is clean, 1 is findings, 2 is a run error. Done when you have the exit code and the findings.
5. **Report.** For each finding, give the path and the fix, or say why it stays. A finding is not a go: file a fix as a task with the `task` skill. Done when each finding has a fix, a task, or a reason.

## Gotchas

- Read the new workspace's `AGENTS.md` before you work in it.
- The plan is create-only. It never changes a file that already exists.
- Keep a check report outside the workspace it checks.

Files in this skill

  • SKILL.md2 KB
  • replays/plan-before-apply.json205 B

Attribution

Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.

Comments

Loading comments…

Newsletter

Get the weekly list of safe new skills, and the ones we pulled.