herald_send

THE RISK

Medium Risk

Why herald_send needs a policy

The tool sends notifications, which is a write operation that creates/transmits data and has side effects (notifications are delivered to recipients). This is not a read operation (it performs an action beyond retrieval), not destructive (notifications can be resent or are not permanent losses), and not financial (no money moves). Write is the appropriate category.

From the tool's definition Tool performs 'Send a notification' action which creates and transmits a message; this is a write operation that modifies external state by delivering a notification through notifykit service.

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

POLICY

How to control herald_send

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

policy.json

{
  "version": "1",
  "default": "deny",
  "tools": {
    "herald_send": {
      "limits": [
        {
          "counter": "herald_send_rate",
          "window": "minute",
          "max": 30,
          "scope": "grant"
        }
      ]
    }
  }
}

herald_send stays usable, but capped — an agent stuck in a loop can't make hundreds of changes a minute. Everything else on the server is denied unless you say otherwise.

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

LIMIT THIS TOOL →

Free to start. No card required.

EXPLORE

Related tools and policies

More Clocktower tools

Execute darkroom_optimize Processes images in a directory using shutterbox image pipeline. Scans the directory for i Write clocktower_init Scaffold a croncall.config.ts file in the target project directory. Write darkroom_init Creates shutterbox.config.ts in a project directory. Scaffolds a starter config with thumb Write flagpost_init Creates flagpost.config.ts in a project directory. Scaffolds a starter config with an exam Write floodgate_init Creates floodgate.config.ts in a project directory. Scaffolds a starter config with defaul Write lockbox_init Initialize lockbox in a project. Creates .vaultbox-key, .secrets/ directory, and updates . Write sifter_init Scaffolds a searchcraft.config.ts in a project directory with a starter search schema. Det Write clocktower_add_job Add a job definition to an existing croncall config file.

All 26 Clocktower tools →

Write tools on other servers

M-Team MCP Server download_torrent YouTube MCP Server generate_video_title Android Forensics ADB MCP Server combine_reports 0xarchive web3_signup

Go deeper

The MCP Attack Database → Documented attack patterns against MCP deployments — and the policies that stop them.
Argument validation → Checking tool call arguments against schema and policy before they run.
Scoped token → Per-person credentials that grant a defined subset of servers and tools.
Rate Limiting MCP Tool Calls: A Practical Guide →
MCP Security: Why Prompt Guardrails Aren't Enough →

FAQ

Questions about herald_send

What does the herald_send tool do? +

Send a notification via notifykit. Returns instructions for configuring providers if the project is not yet set up. It is categorised as a Write tool in the Clocktower MCP Server, which means it can create or modify data. Consider rate limits to prevent runaway writes.

How do I enforce a policy on herald_send? +

Register the Clocktower MCP server in PolicyLayer and add a rule for herald_send: 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 Clocktower. Nothing to install.

What risk level is herald_send? +

herald_send is a Write tool with medium risk. Write tools should be rate-limited to prevent accidental bulk modifications.

Can I rate-limit herald_send? +

Yes. Add a rate_limit block to the herald_send 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 herald_send completely? +

Set action: deny in the PolicyLayer policy for herald_send. 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 herald_send? +

herald_send is provided by the Clocktower MCP server (sathergate/sathergate-toolkit). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.

Enforce policy on every Clocktower tool call.

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

CHECK YOUR STACK →