CVE-2026-26013: LangChain ChatOpenAI SSRF via Image Token Counting
LangChain
LangChain fetched arbitrary image URLs without validation during vision model token counting, leading to blind SSRF. The impact is limited due to timeouts and lack of response return to attackers.
disclosed CVENo confirmed exploitation, so this record is tracked as a disclosed vulnerability and not counted as an incident. It moves into the counts if CISA or the vendor confirms exploitation. Why
- CVSS
- 3.7
- EPSS
- <0.01 · top 63% · 2026-11-13
- fix status
- Fix status unknown
- first report
- 10 Feb 2026
- last update
- 11 Feb 2026
- CVE ids
Timeline · 2
Sources · 2
Advisories and databases
- 2026-02-10CVE-2026-26013: LangChain is a framework for building agents and LLM-powered applications.NVD · nvd.nist.gov/vuln/detail/CVE-2026-26013
- 2026-02-11CVE-2026-26013: LangChain affected by SSRF via image_url token counting in ChatOpenAI.get_num_tokens_from_messagesGitHub Advisory Database · github.com/advisories/GHSA-2g6r-c272-w58r
Titles and links belong to their publishers. The summary above is written by gcdRadar in its own words; data is CC BY-SA 4.0.
Classification
- lane
- Patch now
- AI role
- Attack on AI
- categories
- Infrastructure vulnerability
- severity
- 2 · Disclosed vulnerability
- fix status
- Fix status unknown
- targets
- Framework / tooling
- owasp llm top 10
- LLM03:2025 Supply Chain
- mitre atlas
- no technique for these categories
- dates
- happened 11 Feb 2026 · first reported 10 Feb 2026
- origin
- GitHub Advisory DatabaseVulnerability record from NVD / the GitHub Advisory Database. This product uses the NVD API but is not endorsed or certified by the NVD.
- record
- confidence 0.95 · created 2026-10-18 07:00Z · updated 2026-10-18 10:04Z
Lane, AI role, categories and severity come from the classifier; the OWASP and ATLAS rows are derived from the categories. How these are assigned
Copy a brief
This record as a short Markdown brief, for a ticket, a chat or a coding agent.
the brief as text
# CVE-2026-26013: LangChain ChatOpenAI SSRF via Image Token Counting https://gcdradar.vercel.app/incidents/cve-2026-26013-langchain-chatopenai-ssrf-via-image-token-cou · AIR-2026-0468 · closed 2026-02-10 - Lane: Patch now - Severity: 2 (Disclosed vulnerability) - Fix status: Fix status unknown - First reported: 2026-02-10 (happened 2026-02-11) - AI role: Attack on AI - Products: LangChain - CVE ids: CVE-2026-26013 - CVSS: 3.7 ## Summary LangChain fetched arbitrary image URLs without validation during vision model token counting, leading to blind SSRF. The impact is limited due to timeouts and lack of response return to attackers. ## What to do No action recorded; read the sources below and the vendor's advisory. ## Latest updates - 2026-02-11: CVE-2026-26013: LangChain affected by SSRF via image_url token counting in ChatOpenAI.get_num_tokens_from_messages (https://github.com/advisories/GHSA-2g6r-c272-w58r) - 2026-02-10: CVE-2026-26013: LangChain is a framework for building agents and LLM-powered applications. (https://nvd.nist.gov/vuln/detail/CVE-2026-26013) ## Sources - 2026-02-10 NVD: CVE-2026-26013: LangChain is a framework for building agents and LLM-powered applications. https://nvd.nist.gov/vuln/detail/CVE-2026-26013 - 2026-02-11 GitHub Advisory Database: CVE-2026-26013: LangChain affected by SSRF via image_url token counting in ChatOpenAI.get_num_tokens_from_messages https://github.com/advisories/GHSA-2g6r-c272-w58r Verify against the vendor's advisory before changing production. Data CC BY-SA 4.0, https://gcdradar.vercel.app.
Related
- closed9.8Langflow CVE-2026-27966: Langflow Remote Code Execution in CSV Agent
Patch now · Disclosed · since 27 Feb 2026 · updated 27 Feb 2026 · CVE-2026-27966
AI - closed8.7LangChain family CVE-2024-58340: LangChain MRKLOutputParser regular expression denial-of-service
Patch now · Disclosed · since 12 Jan 2026 · CVE-2024-58340
AI - closed5.1LangChain family CVE-2026-55443: LangChain: Path traversal and sandbox escape in file loaders
Patch now · Disclosed · since 16 Jun 2026 · updated 22 Jun 2026 · CVE-2026-55443
AI - closed8.2LangChain family CVE-2026-44843: LangChain vulnerable to unsafe deserialization
Patch now · Disclosed · since 8 May 2026 · updated 26 May 2026 · CVE-2026-44843
AI - closed3.1LangChain family LangChain langchain-openai TOCTOU DNS rebinding vulnerability
Patch now · Disclosed · since 24 Apr 2026 · CVE-2026-41488
AI - closed6.5LangChain family LangChain URL validation SSRF vulnerability in HTMLHeaderTextSplitter
Patch now · Disclosed · since 24 Apr 2026 · CVE-2026-41481
AI