Create and start a provider-agnostic task, returning a task_id you can track. (v1.0.34) Dispatches the prompt to the provider registered for the given task_type (Codex, Copilot, Claude CLI) and returns immediately with a task_id. Use wait-task to block until the task reaches a terminal state or n...
Risk signalsAccepts file system path (cwd) · High parameter count (13 properties)
Part of the Mcp Codex Worker server.
Free to start. No card required.
AI agents invoke spawn-task to trigger processes or run actions in Mcp Codex Worker. Execute operations can have side effects beyond the immediate call -- triggering builds, sending notifications, or starting workflows. Rate limits and argument validation are essential to prevent runaway execution.
spawn-task can trigger processes with real-world consequences. An uncontrolled agent might start dozens of builds, send mass notifications, or kick off expensive compute jobs. PolicyLayer enforces rate limits and validates arguments to keep execution within safe bounds.
Execute tools trigger processes. Rate-limit and validate arguments to prevent unintended side effects.
{
"version": "1",
"default": "deny",
"tools": {
"spawn-task": {
"limits": [
{
"counter": "spawn-task_rate",
"window": "minute",
"max": 10,
"scope": "grant"
}
]
}
}
}See the full Mcp Codex Worker policy for all 5 tools.
These attack patterns abuse exactly the kind of access spawn-task gives an agent. Each links to the full case and the policy that stops it:
Other execute tools across the catalogue. The same approach applies to each: rate-limit and validate the arguments.
Create and start a provider-agnostic task, returning a task_id you can track. (v1.0.34) Dispatches the prompt to the provider registered for the given task_type (Codex, Copilot, Claude CLI) and returns immediately with a task_id. Use wait-task to block until the task reaches a terminal state or needs input, respond-task to unblock it, and message-task to send follow-ups on the same session. PARALLEL EXECUTION: Spawn multiple tasks in the same message to fan out work — each task runs in its own isolated agent workspace and reports back independently. AFTER SPAWNING: Always follow up with wait-task. The agent may pause almost immediately to request input; the bridge window surfaces that pending question so you can answer it without polling. WRITING A GOOD PROMPT: Name the exact files, functions, or symbols involved, state the expected behavior, and mention anything the agent must NOT touch.. It is categorised as a Execute tool in the Mcp Codex Worker MCP Server, which means it can trigger actions or run processes. Use rate limits and argument validation.
Register the Mcp Codex Worker MCP server in PolicyLayer and add a rule for spawn-task: 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 Codex Worker. Nothing to install.
spawn-task 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 spawn-task 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 spawn-task. 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.
spawn-task is provided by the Mcp Codex Worker MCP server (mcp-codex-worker). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Deterministic rules across all 5 Mcp Codex Worker tools. Per-identity grants. Full audit log. Live in minutes. Nothing to install.
Free to start. No card required.
4,600+ MCP servers and 31,000+ tools scanned and risk-classified.