Searches for records across multiple indexes, deduplicating and reranking the results.
AI agents call cascading-search to retrieve information from Pinecone Developer MCP without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool performs read-only operations: searching, deduplicating, and reranking are all data retrieval and presentation tasks with no side effects on the underlying data. No data is created, modified, deleted, or destroyed. The blast radius of misuse is minimal — an attacker could only retrieve potentially sensitive records they already have query access to, but cannot alter system state or data integrity.
From the tool's definition Tool name contains 'search' and description states it 'Searches for records across multiple indexes, deduplicating and reranking the results' — operations that retrieve and filter data without modification, creation, or deletion.
Documented attack patterns abuse exactly the kind of access cascading-search gives an agent:
PolicyLayer is an MCP gateway — it sits between your AI agents and Pinecone Developer MCP, and nothing reaches the server without passing your rules. This is the rule we recommend for cascading-search:
{
"version": "1",
"default": "deny",
"tools": {
"cascading-search": {}
}
}cascading-search 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.
Searches for records across multiple indexes, deduplicating and reranking the results. It is categorised as a Read tool in the Pinecone Developer MCP MCP Server, which means it retrieves data without modifying state.
Register the Pinecone Developer MCP server in PolicyLayer and add a rule for cascading-search: 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 Pinecone Developer MCP. Nothing to install.
cascading-search 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 cascading-search 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 cascading-search. 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.
cascading-search is provided by the Pinecone Developer MCP server (pinecone-io/pinecone-mcp). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Deterministic rules across all 9 Pinecone Developer MCP tools. Per-identity grants. Full audit log. Live in minutes. Nothing to install.
Free to start. No card required.
9 Pinecone Developer MCP tools catalogued and risk-classified — across an index of 42,500+ MCP servers.