getcursor
Requesting the IP Range of the @docs Crawler for Site-Specific Access Control
Problem Description
When using Cursor's @docs feature, I want to index private documents into Cursor for team use. However, since these documents are private, I need to restrict access to my server resources so that only Cursor's crawler can reach them. Currently, the @docs feature supports indexing public documents via URLs, but for private documents, I need to know the crawler's IP range to configure access control policies on my server (e.g., firewall rules or IP whitelisting) to allow only Cursor's crawler to access these resources.
Feature Request
Could you provide the IP range or IP addresses used by the @docs crawler? My use case is as follows:
Host private documents (e.g., API documentation, internal specifications) on my server. Configure server access controls to allow only IPs from Cursor's crawler to access these documents. Use @docs to index these private documents into Cursor for team use during coding. If the crawler's IP range is dynamic or provided by a third-party service, could you specify the source (e.g., a specific AWS or Google Cloud region) or provide an API to query the latest IP range?
Current Workarounds
I’ve considered the following approaches but would appreciate official guidance:
Guessing the Crawler IP: Observing crawler IPs via server logs, but this is unreliable and not forward-looking. Building a Middle Layer: Converting private documents to Markdown and storing them in a code repository, but this is costly to maintain for large-scale documentation. Reverse Engineering: Attempting to deduce the crawler IP, but this isn’t a secure or efficient solution.
Additional Context
Use Case: Our team has a large number of private documents (e.g., Confluence pages or internal API docs) that we’d like to integrate into Cursor via @docs to improve development efficiency. Related Discussions: I’ve seen similar needs in the Cursor community forum (e.g., Best practices for private documentation indexing?), but no clear solution was provided.
Desired Response
The fixed IP range of the @docs crawler or a method to obtain the dynamic IP range. If this isn’t currently supported, are there plans to offer this feature or an alternative (e.g., supporting authenticated access to private documents) in the future? Thank you for your support! I’m happy to provide more details if needed.
