run_flow · Visual Annotation MCP: Risk & Policy

WHAT IT DOES

What run_flow does on Visual Annotation MCP

AI agents invoke run_flow to trigger actions in Visual Annotation MCP. 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 run_flow needs a policy

The tool executes programmatic flows rather than simply reading or passively inspecting. While it does not inherently delete data (Destructive) or move money (Financial), it performs browser automation and multi-step interactions whose real-world effects are unpredictable without knowing the flow definition. Misuse could trigger unintended purchases, form submissions, or account modifications.

From the tool's definition 'Execute a multi-step interaction flow' indicates the tool runs a sequence of operations whose effects depend on the flow's contents.

Attacks that exploit this kind of access

Recommended rule Ready

Verdict

run_flow Rate-limited

run_flow 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": {
    "run_flow": {
      "limits": [
        {
          "counter": "run_flow_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 Visual Annotation MCP, so you can run agents against it safely.

FAQ

Questions about run_flow

What does the run_flow tool do? +

Execute a multi-step interaction flow (legacy alias). It is categorised as a Execute tool in the Visual Annotation MCP 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_flow? +

Register the Visual Annotation MCP server in PolicyLayer and add a rule for run_flow: 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 Visual Annotation MCP. Nothing to install.

What risk level is run_flow? +

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

Can I rate-limit run_flow? +

Yes. Add a rate_limit block to the run_flow 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_flow completely? +

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

run_flow is provided by the Visual Annotation MCP server (mstocker1/visual_annotation_mcp). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.