Menu
Executes Python code in a temporary sandbox to verify it works before finalizing a mission.
Install with Codex or Claude Copy this prompt, paste it into Codex, Claude, or another assistant, and let it review the skill page and install it for you.
Based on SOC occupation classification
| name | sandbox |
| description | Executes Python code in a temporary sandbox to verify it works before finalizing a mission. |
| compatibility | Requires Python 3.10+ |
| tags | ["coder"] |
Executes a snippet of Python code in an isolated temporary environment. This skill is primarily used by the Coder to verify that a proposed solution or script works as intended before it is delivered to the user or integrated into the core system.
This skill is used by the Coder to:
The sandbox can run in two modes:
SANDBOX_BACKEND=docker or auto with Docker available.The Docker path disables container networking and mounts only the temporary workspace. The fallback path still uses subprocess.run, isolated Python startup, a temporary workspace, and a sanitized environment, but it is not equivalent to full container or WASM isolation.
The execution logic is contained within scripts/sandbox.py. It captures STDOUT and STDERR to provide a full execution report back to the agent for debugging or confirmation.
Create a calendar event by generating an ICS file and opening it in the default calendar application. Use when the user wants to schedule a meeting, appointment, or reminder.
Synthesize the memory timeline into an updated truth summary by merging recent timeline entries with the existing executive summary via the configured LLM. Use to consolidate long-term memory in μZephyr.
Index local files and optionally watch a folder for automatic RAG indexing using ChromaDB and Whoosh. Use when the user wants to make local files searchable or enable hot-watch on a directory.
Create, update, or refine the system prompts for the Agency roles (Supervisor, Researcher, Coder, Reviewer).
Recursively merges .py files from a directory into a single source string.
Search local files using hybrid semantic and keyword retrieval (ChromaDB + Whoosh). Falls back to recursive grep when the index is unavailable. Use when the user wants to find information in their local documents.