SSRF and local file disclosure via unvalidated URLs
- Published
- Last modified
- Where the record stands: published, rejected by its publisher, or superseded by another id.What do these mean? →
- published
- Whether CISA lists it in its Known Exploited Vulnerabilities catalog: proof that attackers are using it.What do these mean? →
- Not listed in CISA KEV
- Fix
- Fixed in 2.1.0
- Whether working attack code is public (Exploit-DB, Metasploit, Nuclei or GitHub). Public code makes attacks easier.What do these mean? →
- None on Exploit-DB, Metasploit or GitHub
Summary
The `webpage` software, in versions before 2.1.0, is affected through its `HTTP::fetch` and `Webpage::from_url` functions, which could let an attacker read local files on the server or make the server send requests to internal services such as cloud metadata endpoints. The impact depends on how the application uses the fetched responses, and it can expose sensitive local data or internal resources.
Summarized by anthropic/claude-haiku-5.5. Verify details against the original sources.
Read the original description
Applications that pass attacker-controlled URLs to `webpage` versions before 2.1.0 may be vulnerable to server-side request forgery (SSRF) or local file disclosure. `HTTP::fetch` and `Webpage::from_url` passed URLs to libcurl without restricting the URL scheme or validating the resolved destination IP address. Redirects were followed without checking whether the new destination was internal. Depending on how the application uses the response, an attacker could: • Read local files accessible to the application process using `file://` URLs. • Cause requests to internal or link-local services (e.g. cloud metadata endpoints at `http://169.254.169.254`) via direct private-IP URLs or by chaining HTTP redirects through an attacker-controlled server. Starting in 2.1.0, `HTTP::fetch` and `Webpage::from_url` reject non-HTTP(S) URL schemes and destinations that resolve to non-public IP addresses, including after redirects. Applications that intentionally fetch trusted internal resources can opt in to the previous behavior with `HTTP::fetch_unrestricted` or `Webpage::from_url_unrestricted`. Reported by Aiden Mohan.
Timeline
- Advisory published
- Record last updated
Affected versions
- webpagecrates.io
From 0.0.0-0 · fixed in 2.1.0