AI agents call check_runner_sandbox to retrieve information from Todos without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool queries the sandbox and trust policy to determine if an action is permitted. It performs a read-only inspection of policy configuration and state, returning a boolean or status result. No data is created, modified, deleted, or executed as a side effect—it merely informs the caller of what would be allowed.
From the tool's definition Tool name 'check_runner_sandbox' and description 'Check whether a local runner action is allowed by sandbox and workspace trust policy' indicate a querying/validation operation.
Attacks that exploit this kind of access
Check whether a local runner action is allowed by sandbox and workspace trust policy. It is categorised as a Read tool in the Todos MCP Server, which means it retrieves data without modifying state.
Register the Todos MCP server in PolicyLayer and add a rule for check_runner_sandbox: 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 Todos. Nothing to install.
check_runner_sandbox is a Read tool with low risk. Read-only tools are generally safe to allow by default.
Yes. Add a rate_limit block to the check_runner_sandbox 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 check_runner_sandbox. 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.
check_runner_sandbox is provided by the Todos MCP server (@hasna/todos). 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 →