AI agents call modelSearch to retrieve information from Runware MCP Server without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
A model search operation retrieves metadata about available models without modifying data or triggering generation/execution. Even though the description is missing, the naming pattern and context of sibling tools (like list and get operations) indicate this is a data retrieval tool. Confidence is moderate due to the empty description.
From the tool's definition Tool name 'modelSearch' indicates a search/query operation; sibling tools include 'listVideoModels' and 'getVideoModelInfo' which are clearly Read operations. The empty description prevents full certainty.
Documented attack patterns abuse exactly the kind of access modelSearch gives an agent:
PolicyLayer is an MCP gateway — it sits between your AI agents and Runware MCP Server, and nothing reaches the server without passing your rules. This is the rule we recommend for modelSearch:
{
"version": "1",
"default": "deny",
"tools": {
"modelSearch": {}
}
}modelSearch is read-only, so it stays allowed — but everything else on the server is denied unless you say otherwise.
Free to start. No card required.
modelSearch. It is categorised as a Read tool in the Runware MCP Server MCP Server, which means it retrieves data without modifying state.
Register the Runware MCP Server MCP server in PolicyLayer and add a rule for modelSearch: 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 Runware MCP Server. Nothing to install.
modelSearch 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 modelSearch 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 modelSearch. 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.
modelSearch is provided by the Runware MCP Server MCP server (runware/mcp-runware). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Start from Runware MCP Server, add the rest of your stack, and see everything your agents can call. Then put policy on all of it.
Free to start. No card required.
11 Runware MCP Server tools catalogued and risk-classified — across an index of 43,000+ MCP servers.