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 agentmods add commands/raine/workmux/sandboxgit clone --depth 1 https://github.com/raine/workmuxWhat 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 | $0.00006 | $0.01958 |
| Opus 5 | $0.00003 | $0.00979 |
| Sonnet 5 | $0.00001 | $0.00392 |
| Haiku 4.5 | $0.00001 | $0.00196 |
Grade A, and why
sandbox 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 2d 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 — 235 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Commands for managing sandbox functionality across container (Docker/Podman/Apple Container) and Lima backends.
Container commands
sandbox build
Build the sandbox container image locally (two-stage: base + agent).
workmux sandbox build
Builds the image locally for the configured agent. This is an alternative to using the pre-built image from ghcr.io/raine/workmux-sandbox. Most users should use workmux sandbox pull instead.
sandbox pull
Pull the latest sandbox image from the container registry.
workmux sandbox pull
Pulls the pre-built image for the configured agent from ghcr.io/raine/workmux-sandbox:{agent}. This is the recommended way to get and update the sandbox image.
sandbox init-dockerfile
Export a customizable Dockerfile for building your own sandbox image.
workmux sandbox init-dockerfile [--force]
Options:
--force- Overwrite existing Dockerfile
Creates a Dockerfile.sandbox in the current directory with the base system setup (Debian, git, workmux) and agent-specific installation (e.g., Claude Code) combined into a single file.
Lima commands
sandbox stop
Stop Lima VMs to free resources.
# Interactive mode - show list and select VM
workmux sandbox stop
# Stop specific VM
workmux sandbox stop <vm-name>
# Stop all workmux VMs
workmux sandbox stop --all
# Skip confirmation prompt
workmux sandbox stop --all --yes
Arguments:
<vm-name>- Name of the VM to stop (optional, conflicts with--all)
Options:
--all- Stop all workmux VMs (those starting withwm-prefix)-y, --yes- Skip confirmation prompt
This command helps you stop running Lima VMs created by workmux to free up system resources. When run without arguments, it shows an interactive list of running workmux VMs for you to choose from. The command will ask for confirmation before stopping any VMs unless --yes is provided.
Notes:
- This command only works with Lima backend and requires
limactlto be installed - Only running VMs are shown in interactive mode
- If a specified VM is already stopped, the command reports this and exits successfully
- Non-interactive environments (pipes, scripts) require
--allor a specific VM name
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.
- 2d ago First seen · 235 lines · 6 tokens per session scan A 25793e244ca0
sandbox is a command published in the GitHub repository raine/workmux (2,351 stars, last pushed today), licensed MIT. It adds 6 tokens to every session and 1,958 once invoked, about $0.0000 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 commands, from other repositories
checklist
Generate a custom checklist for the current feature based on user requirements.
clarify
Identify underspecified areas in the current feature spec by asking up to 5 highly targeted clarification questions and encoding answers back into the spec.
specify
Create or update the feature specification from a natural language feature description.
analyze
Perform a non-destructive cross-artifact consistency and quality analysis across spec.md, plan.md, and tasks.md after task generation.
constitution
Create or update the project constitution from interactive or provided principle inputs.
converge
Assess the current codebase against the feature's spec, plan, and tasks, then append any remaining unbuilt work as new tasks to tasks.md so implement can complete it.