invoke_raw · Claudegram MCP: Risk & Policy

WHAT IT DOES

What invoke_raw does on Claudegram

AI agents invoke invoke_raw to trigger actions in Claudegram. 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.

WHY IT NEEDS A POLICY

Why invoke_raw needs a policy

A 'raw invoke' capability on a protocol client is an Execute-class tool—it triggers external operations (Telegram API calls) whose effects depend entirely on arguments provided by the caller. The lack of description increases concern that this is a low-level escape hatch.

From the tool's definition Tool name 'invoke_raw' on a Telegram MTProto MCP server with empty description strongly suggests direct protocol invocation.

Attacks that exploit this kind of access

Recommended rule Ready

Verdict

invoke_raw Rate-limited

invoke_raw 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.

View as policy code

policy.json

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

RATE-LIMIT THIS TOOL →

Instant setup, no code required.

PolicyLayer is an MCP gateway: every tool call is checked against your rule before it reaches Claudegram, so you can run agents against it safely.

FAQ

Questions about invoke_raw

What does the invoke_raw tool do? +

invoke_raw. It is categorised as a Execute tool in the Claudegram MCP Server, which means it can trigger actions or run processes. Use rate limits and argument validation.

How do I enforce a policy on invoke_raw? +

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

What risk level is invoke_raw? +

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

Can I rate-limit invoke_raw? +

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

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

invoke_raw is provided by the Claudegram MCP server (sanjar-x/claudegram). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.