The local-deep-research AI assistant uses a validate_url function to check URLs before sending requests to prevent SSRF attacks. However, the validation relies on urlparse to extract the host portion, while the actual request is sent using requests.get, which parses URLs differently. This discrepancy allows attackers to craft URLs that bypass validation but are accepted by requests.get, enabling SSRF attacks. The vulnerability affects versions before 1.6.10 and is addressed by correcting the validation logic in version 1.6.10.

An attacker with at least low privileges can exploit this SSRF vulnerability to make the server send unauthorized requests to internal or external resources. The CVSS vector indicates network attack vector, low attack complexity, low privileges required, no user interaction, and a confidentiality impact limited to partial information disclosure. There is no impact on integrity or availability. No known exploits are reported in the wild.

This vulnerability is fixed in local-deep-research version 1.6.10. Users should upgrade to version 1.6.10 or later to remediate this issue. Patch status is not explicitly confirmed by vendor advisory content, but the description states the issue is fixed in 1.6.10. No additional mitigation steps are indicated.