Search bookmarks by title or URL.
AI agents call search_bookmarks to retrieve information from Chromium Sync without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool queries bookmarks without side effects. However, it accesses sensitive personal data (bookmarks) from a user's browser profile, which could expose browsing habits, private websites, or sensitive URLs if misused by an agent. Severity is medium rather than low due to the sensitivity of the data accessed, even though the operation itself is read-only.
From the tool's definition Tool name and description indicate retrieval only: 'Search bookmarks by title or URL.' No modification, deletion, or execution verbs present. Part of a suite that reads browser data from local profile files.
Attacks that exploit this kind of access
Search bookmarks by title or URL. It is categorised as a Read tool in the Chromium Sync MCP Server, which means it retrieves data without modifying state.
Register the Chromium Sync MCP server in PolicyLayer and add a rule for search_bookmarks: 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 Chromium Sync. Nothing to install.
search_bookmarks 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_bookmarks 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_bookmarks. 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_bookmarks is provided by the Chromium Sync MCP server (jaidhyani/chromium-sync-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 →