candidate.evaluate_transaction

What candidate.evaluate_transaction does on Visionmcp

AI agents invoke candidate.evaluate_transaction to trigger actions in Visionmcp. 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 candidate.evaluate_transaction is rated High

This tool runs an automated evaluation process that not only reads/checks data but also executes consequential actions (auto-rejection of regressions). The 'auto-reject' behavior means it has side effects beyond mere reading — it modifies state or outcomes of a transaction. It does not permanently delete data but does execute logic that changes outcomes, placing it in Execute.

From the tool's definition 'Evaluate every mandatory gate atomically and auto-reject any regression' — triggers an automated evaluation pipeline that performs gate checks and auto-rejects (takes action) based on results

Risk signalsBulk/mass operation — affects multiple targets

The rule that runs candidate.evaluate_transaction safely

PolicyLayer is an MCP gateway: it sits between your AI agents and Visionmcp, and checks every tool call against a rule you set before the call runs. Nothing changes on the server itself. For candidate.evaluate_transaction, this is the rule to start with:

{
  "version": "1",
  "default": "deny",
  "tools": {
    "candidate.evaluate_transaction": {
      "limits": [
        {
          "counter": "candidate.evaluate_transaction_rate",
          "window": "minute",
          "max": 10,
          "scope": "grant"
        }
      ]
    }
  }
}

Questions about candidate.evaluate_transaction