run_ai_action

THE RISK

High Risk

Why run_ai_action needs a policy

This tool executes an external operation (an AI Action/Prompt) whose effects depend on which message ID arguments are supplied. The outcome is not guaranteed to be benign—an AI Action could perform various operations on or with the message data.

From the tool's definition Tool description states 'Run an AI Action (Prompt) for a message' and accepts message IDs to execute an AI Action. The word 'Run' combined with 'AI Action (Prompt)' indicates code or command execution triggered by user-supplied arguments (message IDs).

Documented attack patterns abuse exactly the kind of access run_ai_action gives an agent:

POLICY

How to control run_ai_action

PolicyLayer is an MCP gateway — it sits between your AI agents and Carbon Voice, and nothing reaches the server without passing your rules. This is the rule we recommend for run_ai_action:

policy.json

{
  "version": "1",
  "default": "deny",
  "tools": {
    "run_ai_action": {
      "limits": [
        {
          "counter": "run_ai_action_rate",
          "window": "minute",
          "max": 10,
          "scope": "grant"
        }
      ]
    }
  }
}

run_ai_action stays usable, but rate-capped — a runaway agent can't fire it dozens of times a minute. Everything else on the server is denied unless you say otherwise.

Create a free account and register Carbon Voice — nothing to install.
Add this policy — paste it, or build it visually.
Point your MCP client (Claude, Cursor, anything) at your gateway URL.

RATE-LIMIT THIS TOOL →

Free to start. No card required.

EXPLORE

Related tools and policies

More Carbon Voice tools

Destructive delete_folder Delete a folder by its ID. Deleting a folder will also delete nested folders and all the m Execute run_ai_action_for_shared_link Run an AI Action (Prompt) for a shared link. You can run an AI Action for a shared link by Write add_attachments_to_message Add attachments to a message. In order to add attachments to a message, you must provide a Write create_conversation_message Sends a message to an existing conversation or any type with a conversation_id. Write create_direct_message Send a Direct Message (DM) to a User or a Group of Users. Write create_folder Create a new folder. Write create_voicememo_message Create a VoiceMemo Message. In order to create a VoiceMemo Message, you must provide a tra Write move_folder Move a folder by its ID. Move a Folder into another Folder or into a Workspace.

All 28 Carbon Voice tools →

Execute tools on other servers

Android Forensics ADB MCP Server adb_shell_command EMBA-MCP run_emba_scan Frida Game Hacking MCP remove_breakpoint Procmon request_elevation

Go deeper

High-risk tools in Carbon Voice → Every high-risk tool this server exposes, in one view.
All high-risk MCP tools → Every high-risk tool across the catalogue, and the policy pattern that contains them.
The MCP Attack Database → Documented attack patterns against MCP deployments — and the policies that stop them.
Agent sandbox → Bounding what an agent can execute and where the boundary should sit.
Least privilege for MCP → Granting agents the minimum tool surface the task needs.
Rate Limiting MCP Tool Calls: A Practical Guide →
MCP Security: Why Prompt Guardrails Aren't Enough →

FAQ

Questions about run_ai_action

What does the run_ai_action tool do? +

Run an AI Action (Prompt) for a message. You can run an AI Action for a message by its ID or a list of message IDs. It is categorised as a Execute tool in the Carbon Voice MCP Server, which means it can trigger actions or run processes. Use rate limits and argument validation.

How do I enforce a policy on run_ai_action? +

Register the Carbon Voice MCP server in PolicyLayer and add a rule for run_ai_action: 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 Carbon Voice. Nothing to install.

What risk level is run_ai_action? +

run_ai_action is a Execute tool with high risk. Execute tools should be rate-limited and have argument validation enabled.

Can I rate-limit run_ai_action? +

Yes. Add a rate_limit block to the run_ai_action 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.

How do I block run_ai_action completely? +

Set action: deny in the PolicyLayer policy for run_ai_action. 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.

What MCP server provides run_ai_action? +

run_ai_action is provided by the Carbon Voice MCP server (phononx/cv-mcp-server). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.

Enforce policy on every Carbon Voice tool call.

Start from Carbon Voice, add the rest of your stack, and see everything your agents can call. Then put policy on all of it.

CHECK YOUR STACK →