sandbox
SolidProvides read_file/write_file/exec/list_files/read_file/write_file for running process and managing filesystems in the sandbox. Ideal for code testing, file management, and command execution. The sub_claude_agent tool is available for advanced use cases. You MUST load this skill BEFORE use sandbox tools.
Install
Quality Score: 88/100
Skill Content
Details
- Author
- wecode-ai
- Repository
- wecode-ai/Wegent
- Created
- 6 months ago
- Last Updated
- today
- Language
- Python
- License
- Apache-2.0
Integrates with
Similar Skills
Semantically similar based on skill content — not just same category
dev-sandbox
Use when work should run inside a per-branch Docker dev sandbox - initiating one, checking state, attaching over ssh, executing commands or tests inside the Linux clone, refreshing it from origin, and tearing it down. Trigger on "run this in a sandbox", "spin up a sandbox", "sandbox status", "attach to the sandbox", "test this on Linux", "verify cross-platform", "run the suite in the container", working a second feature branch in parallel, or ANY time an in-sandbox result is about to be trusted - the origin-clone caveat here decides whether that result means anything.
sandbox-routing
SELF-CORRECT tool choice. About to Read a long file just to extract one piece of info? Use a sandbox tool that returns only the summary. About to run a long shell command? Same. Sandbox keeps raw output OUT of context — only summaries land. Triggers when about to flood context with large output.
sandboxing-agentic-systems
Contain an agent that runs code or reads untrusted content, layer by layer. Covers OS-level filesystem and network isolation that also catches spawned subprocesses, an egress proxy that checks request provenance, treating tool outputs and fetched pages as prompt-injection vectors, and keeping credentials outside the sandbox behind a proxy. Use this whenever someone designs a sandbox for a coding or computer-use agent, asks how to safely let an agent run shell commands or browse, worries about prompt injection from tool results, needs to limit what an autonomous agent can reach or delete, or asks where an agent's credentials should live. Trigger on "sandbox the agent," "agent runs untrusted code," "prompt injection," "restrict network access," and similar. This is environment containment; deciding which actions need approval before they run is a separate concern (action gating).