Loading vulnerability…
CVE-2025-68665 is a serialization-injection vulnerability in LangChain JS affecting @langchain/core versions before 0.3.80 and 1.1.8, and langchain versions before 0.3.37 and 1.2.3. The vulnerable toJSON() implementation does not escape free-form, user-controlled data in kwargs that contains the internal lc key. Because lc marks LangChain serialized objects, such data can be interpreted as a legitimate LangChain object rather than plain data when deserialized. This can enable secret extraction and unsafe instantiation of classes available through import maps.
Mallory correlates every CVE against your assets, your vendors, and active adversary campaigns. Know which vulnerabilities matter for you, not just which ones are loud.
What it means. What to do now. Patch path, mitigations, and the assume-compromise checklist.
What an attacker gets, and what they’ve been doing with it.
If you can’t patch tonight, do this now.
Patch, then assume compromise.
No public exploits tracked yet. Mallory keeps watching.
No public exploit code observed for this vulnerability.
Products and vendors Mallory has correlated with this vulnerability. Open in Mallory to drill down to specific CPE configurations and version ranges.
Vendor-confirmed product mapping. Mallory continuously reconciles this list against your asset inventory.
13 sources tracked across advisories, community write-ups, and news. New activity surfaces here as Mallory finds it.
JavaScript/TypeScript variant of the LangChain serialization/deserialization issue in `@langchain/core`/`langchain` enabling attacker-controlled `lc` data to be interpreted as internal objects, potentially leading to secret exposure and unsafe object instantiation/code execution paths.
Critical LangChain JavaScript/TypeScript issue tied to unsafe defaults/handling that can enable unauthorized access to environment secrets (notably via `secretsFromEnv`) and related injection-to-secret-theft scenarios.
A related vulnerability to CVE-2025-68664, affecting the JavaScript/TypeScript versions of LangChain (@langchain/core and langchain npm packages), enabling similar risks of secret exfiltration and code execution via unsafe serialization/deserialization.
A similar deserialization vulnerability in LangChainJS (CVE-2025-68665) highlights risks in JavaScript implementations of agentic AI frameworks.
Query your assets running an affected version, and investigate the blast radius.
Every observed campaign linking this CVE to a named adversary.
Malware families riding this exploit, with evidence and IOCs.
YARA, Sigma, Snort, and vendor rules, auto-deployed to your SIEM.
Cross-references every affected SKU, including bundled OEM variants.
Community discussion across Reddit, Mastodon, and other social sources.