Get recommendations before browser automation
AI agents call pre-browse to retrieve information from Claude Flow without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool retrieves recommendations (a Read operation) prior to browser automation, with no side effects or capability to modify, execute, or delete data. The phrasing 'get recommendations' is explicitly a query/retrieval operation. Severity is low because misuse would only expose advice or suggestions, not trigger destructive automation.
From the tool's definition Tool name 'pre-browse' and description 'Get recommendations before browser automation' indicate a read-only advisory function that retrieves or queries recommendations without performing actions.
Attacks that exploit this kind of access
Get recommendations before browser automation. It is categorised as a Read tool in the Claude Flow MCP Server, which means it retrieves data without modifying state.
Register the Claude Flow MCP server in PolicyLayer and add a rule for pre-browse: 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 Claude Flow. Nothing to install.
pre-browse 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 pre-browse 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 pre-browse. 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.
pre-browse is provided by the Claude Flow MCP server (claude-flow). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.