Omnibus LITERAL substring search across Pinako data surfaces — the broad replacement for search_tabs. Bridge-side scan; no LLM. Use this WHENEVER the user names a literal substring and the question may match non-tab nodes (groups, folders, library titles), non-tree surfaces (libraries, bookmarks,...
AI agents call search_pinako to retrieve information from Pinako AI Bridge without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool retrieves and searches existing data (tabs, groups, folders, library titles, bookmarks, notes) without creating, modifying, or deleting anything. It has no side effects beyond returning search results. The emphasis on 'no LLM' and 'literal substring' further confirms it is a pure read operation with minimal security impact.
From the tool's definition Tool performs "LITERAL substring search" across Pinako data surfaces with "Bridge-side scan; no LLM." Description explicitly indicates it retrieves/queries data without modification. No deletion, execution, writing, or financial operations are mentioned.
Attacks that exploit this kind of access
Omnibus LITERAL substring search across Pinako data surfaces — the broad replacement for search_tabs. Bridge-side scan; no LLM. Use this WHENEVER the user names a literal substring and the question may match non-tab nodes (groups, folders, library titles), non-tree surfaces (libraries, bookmarks, notes), or wants exact-tag semantics (default fuzzy substring matching can over-match a tag like. It is categorised as a Read tool in the Pinako AI Bridge MCP Server, which means it retrieves data without modifying state.
Register the Pinako AI Bridge MCP server in PolicyLayer and add a rule for search_pinako: 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 Pinako AI Bridge. Nothing to install.
search_pinako 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 search_pinako 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 search_pinako. 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.
search_pinako is provided by the Pinako AI Bridge MCP server (teleomorph/pinako-mcp). 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 →