mcp-from-openapi from 2.5.0 before 2.9.0 Credential Leak via Cross-Origin Redirects
- 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- Where the record stands: published, rejected by its publisher, or superseded by another id.What do these mean?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 2.9.0
- 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
mcp-from-openapi, a tool that generates MCP servers from OpenAPI specifications, is affected by this vulnerability. An attacker who controls a spec host or can abuse an open redirect could capture users' authentication credentials, such as bearer tokens and cookies, because the software sends them along when following redirects to a different domain.
Summarized by anthropic/claude-haiku-5.5. Verify details against the original sources.
Read the original description
mcp-from-openapi from 2.5.0 before 2.9.0 contains a credential exposure vulnerability that allows attackers to capture Authorization and Cookie headers because safeFetch() forwards them on cross-origin redirects. Attackers controlling a spec host or abusing an open redirect can redirect fromURL() spec requests to their own origin to steal bearer tokens or cookies.
Timeline
- CVE published
- First seen by cve.now
Affected versions
- agentfront mcp-from-openapi
From 2.5.0 · fixed in 2.9.0