HighAI coding agentsBLWK-AI-021

An AI agent's command sandbox is turned off

What this means

An agent's command sandbox is turned off (Codex danger-full-access, Cursor CLI sandbox.mode: disabled, Claude Code's weaker nested sandbox).

How to fix it

Turn the agent's sandbox back on (for example remove sandbox_mode = “danger-full-access” from ~/.codex/config.toml, or sandbox.mode “disabled” from the Cursor CLI config). If one command genuinely needs to escape it, allow that command alone rather than switching the sandbox off for everything.

Check it yourself

Run a scan and look for BLWK-AI-021 in the results:

bulwark scan
bulwark ai scan

Installed the standalone CLI package instead of the app? The command is bulwarkctl.

References

• MITRE ATT&CK T1059
• MITRE ATT&CK T1611

Related checks

• Project-supplied agent hooks run shell commands automatically · BLWK-AI-002
• An agent permission allowlist permits arbitrary command execution · BLWK-AI-006
• An agent is configured to bypass all permission prompts · BLWK-AI-007
• Project MCP servers are set to auto-enable without prompting · BLWK-AI-008
• VS Code chat auto-approve (“YOLO mode”) is enabled · BLWK-AI-009
• VS Code Workspace Trust is disabled · BLWK-AI-010