# cve.now: Langchain-ai

The newest entries naming Langchain-ai as an affected vendor (up to 200), of 55 in all, 0 known exploited. Older ones are in the full catalog at https://cve.now/latest.md.

- [CVE-2026-118668](https://cve.now/CVE-2026-118668.md): LangGraph JavaScript custom authentication may permit cross-user Store access (CVSS 8.6)
- [CVE-2026-118667](https://cve.now/CVE-2026-118667.md): LangGraph: Unencoded identifiers in LangGraph SDK v3 streaming can redirect authenticated GET requests (CVSS 4.3)
- [CVE-2026-115885](https://cve.now/CVE-2026-115885.md): LangGraph CLI prints the registry push token in \`langgraph deploy --verbose\` output (CVSS 4.1)
- [CVE-2026-115883](https://cve.now/CVE-2026-115883.md): LangGraph: Namespace labels containing \`.\` reach other namespaces in Postgres and SQLite stores (CVSS 6.0)
- [CVE-2026-106119](https://cve.now/CVE-2026-106119.md): LangChain: MongoDBChatMessageHistory query injection can allow cross-session access (CVSS 6.0)
- [CVE-2026-105799](https://cve.now/CVE-2026-105799.md): LangChain: RediSearch Filter Injection via Unescaped Tag/Text Values (CVSS 2.3)
- [CVE-2026-104873](https://cve.now/CVE-2026-104873.md): LangGraph SDK custom auth silently ignores actions= on resource decorators (CVSS 7.6)
- [CVE-2026-55253](https://cve.now/CVE-2026-55253.md): LangChain MongoDB: NoSQL Operator Injection in MongoDBSaver.list() leading to cross-tenant data exposure (CVSS 7.7)
- [CVE-2026-55235](https://cve.now/CVE-2026-55235.md): langgraph-api: Relative webhook targets in LangGraph Server can reach in-process routes without authentication (CVSS 5.9)
- [CVE-2026-55236](https://cve.now/CVE-2026-55236.md): langgraph-api: Incomplete assistant authorization in LangGraph Server run creation (CVSS 5.9)
- [CVE-2026-72848](https://cve.now/CVE-2026-72848.md): langchain-community SitemapLoader Does Not Apply restrict\_to\_same\_domain to Nested Sitemap Index Entries, Allowing Server-Side Request Forgery (CVSS 8.6)
- [CVE-2026-71433](https://cve.now/CVE-2026-71433.md): LangGraph: Namespace prefix matching crosses segment boundaries in Postgres and SQLite stores (CVSS 5.3)
- [CVE-2026-48121](https://cve.now/CVE-2026-48121.md): @langchain/langgraph-checkpoint-mongodb: NoSQL parameter injection in MongoDBSaver allows cross-tenant state access (CVSS 6.7)
- [CVE-2026-59152](https://cve.now/CVE-2026-59152.md): Arbitrary server-side file read in LangSmith SDK TracingMiddleware (CVSS 5.0)
- [CVE-2026-14742](https://cve.now/CVE-2026-14742.md): langchain-ai langgraph Task Result Cache \_cache.py \_freeze weak hash (CVSS 2.3)
- [CVE-2026-55443](https://cve.now/CVE-2026-55443.md): LangChain: Path traversal and sandbox escape in LangChain file-search middleware and loaders (CVSS 5.1)
- [CVE-2026-48776](https://cve.now/CVE-2026-48776.md): LangGraph SDK has unsafe URL path construction (CVSS 4.2)
- [CVE-2026-48775](https://cve.now/CVE-2026-48775.md): LangGraph Checkpoint: Unsafe JSON deserialization in checkpoint loading (CVSS 6.8)
- [CVE-2026-45134](https://cve.now/CVE-2026-45134.md): LangSmith Client SDK: Public prompt pull deserializes untrusted manifests without trust boundary warning (CVSS 7.1)
- [CVE-2026-44843](https://cve.now/CVE-2026-44843.md): LangChain: Unsafe deserialization of attacker-controlled LangChain objects through overly broad \`load()\` allowlists (CVSS 8.2)
- [CVE-2026-41488](https://cve.now/CVE-2026-41488.md): angchain-openai: Image token counting SSRF protection can be bypassed via DNS rebinding (CVSS 3.1)
- [CVE-2026-41481](https://cve.now/CVE-2026-41481.md): LangChain: HTMLHeaderTextSplitter.split\_text\_from\_url SSRF Redirect Bypass (CVSS 6.5)
- [CVE-2026-41182](https://cve.now/CVE-2026-41182.md): LangSmith SDK: Streaming token events bypass output redaction (CVSS 5.3)
- [CVE-2026-40190](https://cve.now/CVE-2026-40190.md): LangSmith Client SDKs has Prototype Pollution in langsmith-sdk via Incomplete \`\_\_proto\_\_\` Guard in Internal lodash \`set()\` (CVSS 5.6)
- [CVE-2026-40087](https://cve.now/CVE-2026-40087.md): LangChain has incomplete f-string validation in prompt templates (CVSS 5.3)
- [CVE-2026-34070](https://cve.now/CVE-2026-34070.md): LangChain Core has Path Traversal vulnerabilites in legacy \`load\_prompt\` functions (CVSS 7.5)
- [CVE-2026-28277](https://cve.now/CVE-2026-28277.md): LangGraph: Unsafe msgpack deserialization in LangGraph checkpoint loading (CVSS 6.8)
- [CVE-2026-25750](https://cve.now/CVE-2026-25750.md): LangSmith Studio has URL Parameter Injection Vulnerability that Enables Token Theft via Malicious baseUrl (CVSS 8.5)
- [CVE-2026-27795](https://cve.now/CVE-2026-27795.md): LangChain Community: redirect chaining can lead to SSRF bypass via RecursiveUrlLoader (CVSS 4.1)
- [CVE-2026-27794](https://cve.now/CVE-2026-27794.md): LangGraph: BaseCache Deserialization of Untrusted Data may lead to Remote Code Execution (CVSS 6.6)
- [CVE-2026-27022](https://cve.now/CVE-2026-27022.md): RediSearch Query Injection in @langchain/langgraph-checkpoint-redis (CVSS 6.5)
- [CVE-2026-26019](https://cve.now/CVE-2026-26019.md): @langchain/community affected by SSRF Bypass in RecursiveUrlLoader via insufficient URL origin validation (CVSS 4.1)
- [CVE-2026-26013](https://cve.now/CVE-2026-26013.md): LangChain affected by SSRF via image\_url token counting in ChatOpenAI.get\_num\_tokens\_from\_messages (CVSS 3.7)
- [CVE-2026-25528](https://cve.now/CVE-2026-25528.md): LangSmith Client SDK Affected by Server-Side Request Forgery via Tracing Header Injection (CVSS 5.8)
- [CVE-2024-58340](https://cve.now/CVE-2024-58340.md): LangChain \<= 0.3.1 MRKLOutputParser ReDoS (CVSS 8.7)
- [CVE-2025-68665](https://cve.now/CVE-2025-68665.md): LangChain serialization injection vulnerability enables secret extraction (CVSS 8.6)
- [CVE-2025-68664](https://cve.now/CVE-2025-68664.md): LangChain serialization injection vulnerability enables secret extraction in dumps/loads APIs (CVSS 9.3)
- [CVE-2025-67644](https://cve.now/CVE-2025-67644.md): LangGraph SQLite Checkpoint is vulnerable to SQL Injection via metadata filter key in checkpointer list method (CVSS 7.3)
- [CVE-2025-65106](https://cve.now/CVE-2025-65106.md): LangChain Vulnerable to Template Injection via Attribute Access in Prompt Templates (CVSS 8.3)
- [CVE-2025-64439](https://cve.now/CVE-2025-64439.md): LangGraph Checkpoint affected by RCE in "json" mode of JsonPlusSerializer (CVSS 7.4)
- [CVE-2025-64104](https://cve.now/CVE-2025-64104.md): LangGraph SQLite Checkpoint Filter Key SQL Injection POC for SqliteStore (CVSS 7.3)
- [CVE-2025-8709](https://cve.now/CVE-2025-8709.md): SQL Injection in langchain-ai/langchain (CVSS 7.3)
- [CVE-2025-6985](https://cve.now/CVE-2025-6985.md): XXE Vulnerability in langchain-ai/langchain (CVSS 7.5)
- [CVE-2025-6984](https://cve.now/CVE-2025-6984.md): Sensitive Information Disclosure Due to Insecure XML Parsing in langchain-ai/langchain (CVSS 7.5)
- [CVE-2025-2828](https://cve.now/CVE-2025-2828.md): SSRF Vulnerability in RequestsToolkit in langchain-ai/langchain (CVSS 8.4)
- [CVE-2024-10940](https://cve.now/CVE-2024-10940.md): Exposure of Sensitive System Information via ImagePromptTemplate in langchain-ai/langchain (CVSS 5.3)
- [CVE-2024-8309](https://cve.now/CVE-2024-8309.md): SQL Injection in langchain-ai/langchain (CVSS 4.9)
- [CVE-2024-7042](https://cve.now/CVE-2024-7042.md): Prompt Injection in langchain-ai/langchainjs Leading to SQL Injection (CVSS 4.9)
- [CVE-2024-7774](https://cve.now/CVE-2024-7774.md): Path Traversal in langchain-ai/langchainjs (CVSS 6.5)
- [CVE-2024-5998](https://cve.now/CVE-2024-5998.md): Deserialization of Untrusted Data in langchain-ai/langchain (CVSS 5.2)
- [CVE-2024-2965](https://cve.now/CVE-2024-2965.md): Denial-of-Service in LangChain SitemapLoader in langchain-ai/langchain (CVSS 4.2)
- [CVE-2024-3095](https://cve.now/CVE-2024-3095.md): SSRF in Langchain Web Research Retriever in langchain-ai/langchain (CVSS 4.8)
- [CVE-2024-3571](https://cve.now/CVE-2024-3571.md): Path Traversal in langchain-ai/langchain (CVSS 6.5)
- [CVE-2024-1455](https://cve.now/CVE-2024-1455.md): Billion Laughs Attack leading to DoS in langchain-ai/langchain (CVSS 5.9)
- [CVE-2024-0243](https://cve.now/CVE-2024-0243.md): Server-side Request Forgery In Recursive URL Loader (CVSS 3.7)

---

Source: cve.now
Built by [Sazabi](https://www.sazabi.com/?utm_source=cve-now&utm_medium=markdown), AI-native observability and security: automatic issue detection, plain-language log search and live dashboards.
Page: https://cve.now/vendors/langchain-ai
Site guide for agents: https://cve.now/llms.txt
Uses NVD data; not endorsed by NVD.