capture_observation · Metis Public Health MCP: Risk & Policy

WHAT IT DOES

What capture_observation does on Metis Public Health

AI agents use capture_observation to create or update resources in Metis Public Health — usually the action step of a workflow, after the agent has gathered context. Every call changes real data in your Metis Public Health environment.

WHY IT NEEDS A POLICY

Why capture_observation needs a policy

The name implies creating or saving a new observation entry, which is a Write operation. However, the empty description significantly lowers confidence. Given sibling tools like 'add_journal_entry' and 'add_memory_entry' that follow a similar 'add/capture' pattern for writing data, this tool likely stores an observation in the local library.

From the tool's definition Tool name 'capture_observation' suggests recording/storing an observation; description is empty and uninformative.

Attacks that exploit this kind of access

Recommended rule Ready

Verdict

capture_observation Rate-limited

capture_observation 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.

View as policy code

policy.json

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

LIMIT THIS TOOL →

Instant setup, no code required.

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

FAQ

Questions about capture_observation

What does the capture_observation tool do? +

capture_observation. It is categorised as a Write tool in the Metis Public Health MCP Server, which means it can create or modify data. Consider rate limits to prevent runaway writes.

How do I enforce a policy on capture_observation? +

Register the Metis Public Health MCP server in PolicyLayer and add a rule for capture_observation: 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 Metis Public Health. Nothing to install.

What risk level is capture_observation? +

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

Can I rate-limit capture_observation? +

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

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

capture_observation is provided by the Metis Public Health MCP server (sveritg/metis_ph). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.