AI agents invoke task_submit to trigger actions in Willow. 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 submits tasks for execution in a sandboxed environment. Despite the sandbox (bubblewrap), it triggers external code/command execution whose effects depend on the submitted task arguments. Execution of arbitrary tasks carries high blast radius risk even within a sandbox, as sandbox escapes or resource exhaustion are possible.
From the tool's definition 'Submit a task to the Kart sandboxed execution queue. Kart runs the task in a bubblewrap'
Attacks that exploit this kind of access
Submit a task to the Kart sandboxed execution queue. Kart runs the task in a bubblewrap. It is categorised as a Execute tool in the Willow MCP Server, which means it can trigger actions or run processes. Use rate limits and argument validation.
Register the Willow MCP server in PolicyLayer and add a rule for task_submit: 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 Willow. Nothing to install.
task_submit 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 task_submit 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 task_submit. 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.
task_submit is provided by the Willow MCP server (rudi193-cmd/willow-mcp). 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 →