Exploitation observed; activity peaked at 4 mentions and remains active
Recommended action window: Immediate (within 24h)
NVD description
LangChain is a framework for building agents and LLM-powered applications. Prior to version 1.2.22, multiple functions in langchain_core.prompts.loading read files from paths embedded in deserialized config dicts without validating against directory traversal or absolute path injection. When an application passes user-influenced prompt configurations to load_prompt() or load_prompt_from_config(), an attacker can read arbitrary files on the host filesystem, constrained only by file-extension checks (.txt for templates, .json/.yaml for examples). This issue has been patched in version 1.2.22.
Priority
MEDIUM
Exploitation
ACTIVE
PoC
YES
Patch
AVAILABLE
Momentum
STABLE
If you run products in this scope, you should treat this CVE as relevant to your environment.
| Date | Total | Labels |
|---|
| 2026-03-27 | 1 | Disclosure1 |
| 2026-03-28 | 1 | Disclosure1 |
| 2026-03-30 | 2 | Disclosure2 |
| 2026-03-31 | 4 | Disclosure1General2Patch1 |
| 2026-04-04 | 1 | Patch1 |
| 2026-04-06 | 1 | Active Exploitation1 |
| 2026-04-17 | 1 | Disclosure1 |
| 2026-04-25 | 1 | Patch1 |
| 2026-05-07 | 1 | General1 |
| 2026-05-08 | 1 | Patch1 |
| 2026-06-20 | 1 | Disclosure1 |
GitHub - Rickidevs/CVE-2026-34070: I Found a Zero-Day Vulnerability in langchain — Here’s How It Went · GitHub - https://github.com/Rickidevs/CVE-2026-34070
Post summary
The article announces the discovery of a zero‑day vulnerability in langchain and hosts a GitHub repository that likely contains PoC code, but offers no evidence of active exploitation, patches, or detailed technical specifics.
🚨 LIVE HIJACK ALERT — CVE-2026-34070. CVSS 7.5. langchain prompt loading reads user-controlled file paths without validation. attackers gain arbitrary file read access through deserialized configs. investigating. 🧵
Post summary
CVE‑2026‑34070 in langchain allows arbitrary file read via deserialized configs; active exploitation is currently occurring and under investigation.
6/ 3. LangChain-Core Secret Exfiltration (CVE-2026-34070) • The Bug: Path traversal inside the legacy prompt-loading API (load_prompt()) • The Impact: If an attacker can manipulate configuration paths, they can force the server to read arbitrary local files. Making .env open
Post summary
The text discloses a path traversal vulnerability (CVE‑2026‑34070) in LangChain‑Core’s legacy prompt loading API, explaining its potential to read arbitrary files. No PoC, exploit, patch, or active exploitation evidence is provided.
緊急：LangChain Coreのload_prompt()系APIにパストラバーサル（CVE-2026-34070、CVSS 7.5）。週2,300万DLライブラリの脆弱性で、APIキー・DB認証情報を含む任意ファイルが読み取り可能。マルチテナントLLMアプリを運用中なら v1.2.22 への即時アップデートを。 #AI駆動開発
Post summary
An advisory alerts that LangChain Core’s load_prompt() API contains a path‑traversal vulnerability (CVE‑2026‑34070, CVSS 7.5) allowing arbitrary file read, and it recommends an immediate update to version 1.2.22.
March 2026: Three vulns in LangChain/LangGraph disclosed. CVE-2026-34070 (7.5): Arbitrary file access CVE-2025-68664 (9.3): API key leakage CVE-2025-67644 (7.3): SQL injection 9 million LangGraph downloads that week. How many organizations realized their agents compromised?
Post summary
Three CVEs in LangChain/LangGraph were disclosed in March 2026—covering arbitrary file access, API key leakage, and SQL injection—without mentioning active exploitation, patches, or PoC.
The 3 CVEs: - CVE-2026-34070: Path Traversal (CVSS 7.5) - reads your filesystem - CVE-2025-68664: Deserialization Injection (CVSS 9.3 Critical) - RCE on your server - CVE-2025-67644: SQL Injection (CVSS 7.3) - exposes conversation history Patched in langchain-core 1.2.22.
Post summary
Three CVEs (CVE‑2026‑34070, CVE‑2025‑68664, CVE‑2025‑67644) are highlighted with technical details and have been addressed by updating langchain‑core to version 1.2.22.
Three vulnerabilities in LangChain & LangGraph: path traversal flaw (CVE-2026-34070, CVSS 7.5) exposed files, deserialization bug (CVE-2025-68664, CVSS 9.3) leaked API keys. #security #LangChain #LangGraph
Post summary
The post announces two CVEs for LangChain & LangGraph, detailing a path traversal flaw that exposes files and a deserialization bug that leaks API keys, each accompanied by CVSS scores.
CVE-2026-34070: Path traversal via prompt templates. Load a crafted prompt, read arbitrary files. Docker configs, credentials, source code — all accessible.
Post summary
CVE-2026-34070 is a path traversal flaw permitting attackers to read arbitrary files through crafted prompt templates, exposing Docker configuration files, credentials, and source code.
プログラミングニュース 2026年第19週を公開しました。 今週のハイライト： **セキュリティ（要対応）** - Next.js 16.2.5 — 高危険度CVE 6件修正（DoS・SSRFなど） - PHP 8.2〜8.5 全系統 セキュリティリリース（CVE 8件） - LangChain CVE-2026-34070 パストラバーサル修正 - Vaultwarden 1.36.0 — SSO CSRF修正 **言語** - Node.js 26.0.0 メジャーリリース — Temporal API がデフォルト有効に - Python 3.15.0 beta 1 / 3.14.5 RC リリース - Rust — Outreachy 2026 参加・nvptx64ターゲット要件引き上げ - Kotlin 2.4.0-Beta2 / LinkedIn Learning認定資格リリース **フレームワーク** - React 19.2.6/19.1.7/19.0.6 — Server Components型強化 - React Native 0.86.0-rc.0 / 0.85.3 - Vue.js 3.6.0-beta.11 / 3.5.34 - Laravel v13.8.0 **OSS注目** - Ollama v0.23.0 — Claude Desktop サポート追加 - Outline v1.7.1 — MCP強化（削除ツール追加） - Misskey 2026.5.1 / Strapi v5.45.0 / n8n 2.19.5 **ライブラリ** - axios v1.16.0 — QUERY メソッド対応 - Tokio 1.52.2 / Zod v4.4.3 / Tauri v2.11.1 動画はこちら → https://youtu.be/atbUFUIuhWY #プログラミング #JavaScript #NodeJS #PHP #Rust #NextJS #OSS #個人開発
Post summary
The announcement reports several security patches, including fixes for multiple CVEs in Next.js, PHP, LangChain, and Vaultwarden, providing technical details of the vulnerabilities addressed.
CVE-2026-34070 in `langchain-core` makes a good case for vendoring specific LLM client code instead of pulling in large frameworks. Less surface area, fewer surprises. https://github.com/langchain-ai/langchain/releases/tag/langchain-core%3D%3D0.3.86 https://t.co/D2fusMacvR
Post summary
The tweet notes CVE‑2026‑34070 in langchain‑core and links to a GitHub release, but provides no technical or exploit details.
3 serious CVEs just dropped for LangChain + LangGraph: - CVE-2025-68664 (CVSS 9.3): API keys & env secrets leak via unsafe deserialization - CVE-2026-34070 (CVSS 7.5): Path traversal -- arbitrary file read - CVE-2025-67644 (CVSS 7.3): SQL injection into conversation history 60M downloads/week. The core of your AI stack is now an attack surface. Patch now: v1.2.22+ / v0.3.81+ / langgraph-checkpoint-sqlite v3.0.1+ #AISecurity #LangChain #LLMSecurity
Post summary
Three high‑severity CVEs affecting LangChain/LangGraph are disclosed with detailed technical info and immediate patch versions supplied.
⚡ New CVE Alert: CVE-2026-34070 📊 Severity: 7.5 🚨 Risk Level: High 🧩 Affects: Multiple / Unspecified Products Reference: https://nvd.nist.gov/vuln/detail/CVE-2026-34070 #CVE-2026-34070 #CVE #High #CyberSecurity #InfoSec https://t.co/secVhVb56x
Post summary
The tweet announces CVE‑2026‑34070 with a high severity rating but provides no technical, exploit, or mitigation details.
CVE-2026-34070 LangChain is a framework for building agents and LLM-powered applications. Prior to version 1.2.22, multiple functions in langchain_core.prompts.loading read files fr… https://www.cve.org/CVERecord?id=CVE-2026-34070
Post summary
The post references CVE-2026-34070 with a vague description of functions that read files, but provides no PoC, exploit, patch, or evidence of active exploitation.
🚨 CYBERDUDEBIVASH SENTINEL APEX ALERT 🚨 Threat: CVE-2026-34070 - LangChain Core has Path Traversal vulnerabilites in legacy `load_prompt` functions Intel Report: https://ift.tt/jfizX1W
Post summary
The alert announces a new CVE-2026-34070 affecting LangChain Core, describing Path Traversal vulnerabilities in legacy load_prompt functions; it lacks evidence of PoC, exploit code, or active attacks, and no patches or workarounds are provided.
Three new vulnerabilities (including CVE-2026-34070) have been disclosed that expose local files, secrets, and backend databases via SSRF . The models aren't the only target—the frameworks connecting them to your data are the soft underbelly. Details: https://hubs.ly/Q048SgSL0
Post summary
The post announces the disclosure of three SSRF vulnerabilities, including CVE-2026-34070, exposing local files, secrets, and database information, but provides no PoC, exploitation details, patch information, or debunking claims.
1 of 1 entries
| Part | Vendor | Product | Version | Target SW | Target HW |
|---|---|---|---|---|---|
| App | langchain | langchain_core | - | python | - |