Score multiple AI outputs against the same rubric in parallel and return aggregate statistics (mean, min, max, std dev, outliers). Useful for comparing prompt variations or model outputs at scale.
AI agents call eval_batch to retrieve information from Scan Your Ai Toolkit without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool reads and evaluates AI outputs, computing aggregate statistics. It does not modify, delete, execute, or trigger any external operations — it purely analyzes and returns computed metrics. The main risk is limited to exposure of potentially sensitive AI outputs fed into it, keeping severity low.
From the tool's definition Score multiple AI outputs against the same rubric in parallel and return aggregate statistics (mean, min, max, std dev, outliers)
Attacks that exploit this kind of access
Score multiple AI outputs against the same rubric in parallel and return aggregate statistics (mean, min, max, std dev, outliers). Useful for comparing prompt variations or model outputs at scale. It is categorised as a Read tool in the Scan Your Ai Toolkit MCP Server, which means it retrieves data without modifying state.
Register the Scan Your Ai Toolkit MCP server in PolicyLayer and add a rule for eval_batch: 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 Scan Your Ai Toolkit. Nothing to install.
eval_batch is a Read tool with low risk. Read-only tools are generally safe to allow by default.
Yes. Add a rate_limit block to the eval_batch 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.
Set action: deny in the PolicyLayer policy for eval_batch. 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.
eval_batch is provided by the Scan Your Ai Toolkit MCP server (sakthivelchan89/scan_your_ai_toolkit). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Every MCP server has a record like this.
Type a name, get the same breakdown: verified identity, auth posture, risk grade, capabilities, recommended policy.
Teams ship this data inside their own products. See what a licence covers →