Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add ibrews/claude-fleet --skill relocate-claude-vmgit clone --depth 1 https://github.com/ibrews/claude-fleetWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/ibrews/claude-fleet/relocate-claude-vm)<a href="https://agentmods.dev/skills/ibrews/claude-fleet/relocate-claude-vm"><img src="https://agentmods.dev/badge/skills/ibrews/claude-fleet/relocate-claude-vm/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/ibrews/claude-fleet/relocate-claude-vm"><img src="https://agentmods.dev/badge/skills/ibrews/claude-fleet/relocate-claude-vm.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00098 | $0.01582 |
| Opus 5 | $0.00049 | $0.00791 |
| Sonnet 5 | $0.00020 | $0.00316 |
| Haiku 4.5 | $0.00010 | $0.00158 |
Grade A, and why
relocate-claude-vm scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 12d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 100 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Relocate Claude Desktop's Computer Use VM bundle
The Claude Desktop app downloads a multi-GB sandboxed VM image to power its local Computer Use / code-execution features. The app provides no in-app setting to relocate it. On a system drive that's already tight, this bundle can be the single biggest space consumer — bigger than the user's Documents folder.
This skill moves the bundle to a target drive and leaves a directory junction (Windows) or symbolic link (macOS — not yet implemented) at the original location, so Claude Desktop sees the same path it expects.
When to use this skill
- A disk scan shows
C:\Users\<user>\AppData\Roaming\Claude\vm_bundles\is a top consumer (typically 11-13 GB) - The system drive is below 15% free and you've already cleared regenerable caches (npm, pip, Windows Update, temp)
- The user has at least 25 GB free on another local NTFS volume (need room for the move + headroom)
Do not use this skill when:
- Claude Desktop is mid-update — finish the update first; MSIX state during an update is brittle
- The VHDX is currently in use by an active Computer Use session — the move will fail (the script checks for this and aborts cleanly)
- The destination is a network drive, removable drive, or non-NTFS volume — VHDX-backed VMs need local fixed NTFS storage
What the bundle actually contains
Inside %APPDATA%\Claude\vm_bundles\claudevm.bundle\ you'll typically find:
| File | Typical size | Purpose |
|---|---|---|
rootfs.vhdx |
~9 GB | The mounted VM root filesystem image (QEMU VHDX v2 format — magic bytes vhdxfile) |
sessiondata.vhdx |
~0.5 GB | Per-session writable disk |
smol-bin.vhdx |
~35 MB | Small helper image |
initrd, vmlinuz |
~180 MB combined | Linux kernel + initial ramdisk |
*.zst archives |
~2.4 GB | Compressed source bundles (used to re-extract the VHDX if corrupted) |
.*.origin files |
tiny | Manifest sidecars referencing the source archives |
The conversation history people often assume is in this folder is not here — Claude Desktop conversations live server-side at Anthropic, with only a small (~250 MB) Electron cache locally in Cache, Code Cache, etc.
What ships with it
2 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 12d ago First seen · 100 lines · 98 tokens per session scan A 39d5214eebd3
relocate-claude-vm is a skill published in the GitHub repository ibrews/claude-fleet (22 stars, last pushed 9d ago), licensed MIT. It adds 98 tokens to every session and 1,582 once invoked, about $0.0005 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
Subagent Driven Development
Orchestrate specialized, autonomous AI agents to execute parallel subtasks with strict boundaries.
taiyi-ui-design
A design-planning guide for describing how an application's user interface should look and behave. It produces a UI-DESIGN.md document covering layouts, components, interactions, accessibility, and error states.
taiyi-evolve
A workflow skill that compares the implemented code with the frozen design after development and testing. It records architecture changes and proposes updates to DESIGN.md, the document describing the intended system structure.
taiyi-diagram-c4
A code-scanning tool that builds C4 architecture documents from a repository. It separates facts observed in the code from conclusions inferred about the design and uses Mermaid diagrams as the source format.
shogun-screenshot
A screenshot tool for getting images from a computer or web page and then cropping, resizing, or masking sensitive information. Playwright is a browser-automation tool used here to capture web pages.
config-validator
Validate AIWG configuration files and project setup for correctness and completeness.