Run a free-form local shell command from a startup-configured unrestricted runner root after explicit local session approval. This is intentionally dangerous: the configured path is an approval scope, not an operating-system sandbox, and commands are not allowlisted. 1Password secrets are not inj...
AI agents invoke op_unrestricted_run to trigger actions in Mcp 1password. What it does depends on the arguments the agent supplies, and its effects often reach beyond the immediate call — builds kicked off, notifications sent, workflows started.
This tool executes arbitrary shell commands on the local system without allowlisting or sandboxing protections. While it requires explicit local session approval, the description emphasizes it is 'intentionally dangerous' and does not restrict what commands can be run. Arbitrary command execution represents the highest Execute severity because an AI agent could run destructive, exfiltrative, or malicious commands.
From the tool's definition Tool name contains 'unrestricted_run'; description states 'Run a free-form local shell command' and explicitly notes 'This is intentionally dangerous' with 'commands are not allowlisted'. The tool permits arbitrary shell execution with minimal restrictions.
Risk signalsAdmin/system-level operation
Attacks that exploit this kind of access
Run a free-form local shell command from a startup-configured unrestricted runner root after explicit local session approval. This is intentionally dangerous: the configured path is an approval scope, not an operating-system sandbox, and commands are not allowlisted. 1Password secrets are not injected; prefer op_script_run for secret-consuming commands. If returnOutput=true is requested without plaintext acknowledgement, execution is skipped and the required acknowledgement is returned. It is categorised as a Execute tool in the Mcp 1password MCP Server, which means it can trigger actions or run processes. Use rate limits and argument validation.
Register the Mcp 1password MCP server in PolicyLayer and add a rule for op_unrestricted_run: allow, deny, rate-limit, or require approval. Point your MCP client at the PolicyLayer proxy URL and the rule is enforced on every call, before it reaches Mcp 1password. Nothing to install.
op_unrestricted_run is a Execute tool with high risk. Execute tools should be rate-limited and have argument validation enabled.
Yes. Add a rate_limit block to the op_unrestricted_run rule in your PolicyLayer policy. For example, setting max: 10 and window: 60 limits the tool to 10 calls per minute. Rate limits are tracked per agent session and reset automatically.
Set action: deny in the PolicyLayer policy for op_unrestricted_run. The AI agent will receive a policy violation error and cannot call the tool. You can also include a reason field to explain why the tool is blocked.
op_unrestricted_run is provided by the Mcp 1password MCP server (kefapps/onepassword-mcp-codex). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Every MCP server has a record like this.
Type a name, get the same breakdown: verified identity, auth posture, risk grade, capabilities, recommended policy.
Teams ship this data inside their own products. See what a licence covers →