gha-lab-61c59f4acb

Artefacto de investigación automatizado — no es el proyecto original. Este repositorio es un laboratorio desechable construido por un sistema automatizado para una tesis de maestría en Université Laval sobre la reproducción de vulnerabilidades publicadas en flujos de trabajo de GitHub Actions. Es...

cve-2025-49131-poc

CVE-2025-49131 - FastGPT 沙箱逃逸 POC root@kitploit: ███████╗ █████╗ ███████╗████████╗ ██████╗ ██████╗ ████████╗ ██╔════╝██╔══██╗██╔════╝╚══██╔══╝██╔════╝ ██╔══██╗╚══██╔══╝ █████╗ ███████║███████╗ ██║ ██║ ███╗██████╔╝ ██║ ██╔══╝ ██╔══██║╚════██║ ██║ ██║ ██║██╔═══╝ ██║ ██║ ██║ ██║███████║ ██║...

fastgpt-sandbox-audit

fastgpt-sandbox-audit Herramienta de auditoría de cadenas de escape del sandbox de Python de FastGPT. El nodo de "ejecución de código" del flujo de trabajo de FastGPT envía el código Python del usuario a un sandbox independiente para su ejecución. La protección del sandbox en v4.14.8 es una capa...

CVE-2026-79483-FastGPT-NoSQL-Injection

CVE-2026-79483-FastGPT-NoSQL-Injection English | 中文 Resumen de la Vulnerabilidad...

your-bot-my-inbox-cve-2026-68929-fastgpt-unauthenticated-wechat-channel-hijack

CVE-2026-68929 — FastGPT WeChat Channel Hijack PoC Unauthenticated, cross-tenant takeover of FastGPT's WeChat iLink share channel, keyed only on the publicshareId...

CVE-2026-79483

FastGPT Community Edition 4.10.0 through 4.14.0 are vulnerable to a NoSQL injection in the POST /api/core/chat/getHistories endpoint. An unauthenticated attacker can inject malicious NoSQL operators via crafted JSON payloads to bypass authorization checks, resulting in unauthorized access to chat...

CVE-2026-79483

FastGPT Community Edition 4.10.0 through 4.14.0 are vulnerable to a NoSQL injection in the POST /api/core/chat/getHistories endpoint. An unauthenticated attacker can inject malicious NoSQL operators via crafted JSON payloads to bypass authorization checks, resulting in unauthorized access to chat...

EUVD-2026-68876

FastGPT Community Edition 4.10.0 through 4.14.0 are vulnerable to a NoSQL injection in the POST /api/core/chat/getHistories endpoint. An unauthenticated attacker can inject malicious NoSQL operators via crafted JSON payloads to bypass authorization checks, resulting in unauthorized access to chat...

CVE-2026-79483

FastGPT Community Edition 4.10.0 through 4.14.0 are vulnerable to a NoSQL injection in the POST /api/core/chat/getHistories endpoint. An unauthenticated attacker can inject malicious NoSQL operators via crafted JSON payloads to bypass authorization checks, resulting in unauthorized access to chat...

CVE-2026-79483

FastGPT Community Edition (versions 4.10.0 through 4.14.0) is affected by a NoSQL injection in the POST /api/core/chat/getHistories endpoint. An unauthenticated attacker can inject malicious NoSQL operators via crafted JSON payloads to bypass authorization checks, gaining unauthorized access to c...

CVE-2026-68929

FastGPT is an open-source LLM platform for building AI applications on a knowledge base. In versions prior to 4.15.2, the WeChat iLink share-channel endpoints authorize requests using only the public shareId, with no authenticated identity or team-ownership check. As a result, an unauthenticated...

CVE-2026-68929 FastGPT: Unauthenticated WeChat channel hijack and denial of service via shareId-only authorization

FastGPT is an open-source LLM platform for building AI applications on a knowledge base. In versions prior to 4.15.2, the WeChat iLink share-channel endpoints authorize requests using only the public shareId, with no authenticated identity or team-ownership check. As a result, an unauthenticated...

CVE-2026-68929

FastGPT , an open-source LLM platform, is vulnerable to unauthenticated WeChat channel hijack and denial of service in versions prior to 4.15.2 . The WeChat (iLink) share-channel endpoints authorize requests using only the publicly exposed shareId , with no authenticated identity or team-ownershi...

CVE-2026-68929 FastGPT: Unauthenticated WeChat channel hijack and denial of service via shareId-only authorization

FastGPT is an open-source LLM platform for building AI applications on a knowledge base. In versions prior to 4.15.2, the WeChat iLink share-channel endpoints authorize requests using only the public shareId, with no authenticated identity or team-ownership check. As a result, an unauthenticated...

PT-2026-82775

Name of the Vulnerable Software and Affected Versions FastGPT versions prior to 4.15.2 Description WeChat iLink share-channel endpoints authorize requests using only the public shareId without verifying authenticated identity or team ownership. Because the shareId is exposed in shared chat URLs,...

Exploit for Incomplete List of Disallowed Inputs in Fastgpt

fastgpt-sandbox-audit FastGPT Python sandbox escape chain aud...

CVE-2026-61643

FastGPT is a knowledge-based AI application platform. From 4.14.17 until 4.15.0-beta5, an authenticated FastGPT user can save a workflow node that points to another user's private HTTP toolset by using a crafted saved tool id such as http-/. The normal toolset routes deny access, but the workflow...

CVE-2026-61643

CVE-2026-61643 affects FastGPT (versions 4.14.17 through 4.15.0-beta5). An authenticated user can save a workflow node that points to another user's private HTTP toolset by crafting a saved tool id (e.g., http-/). While normal toolset routes deny access, the workflow save and runtime path did not...

CVE-2026-61643 FastGPT: workflow runtime can execute another user's private HTTP toolset

FastGPT is a knowledge-based AI application platform. From 4.14.17 until 4.15.0-beta5, an authenticated FastGPT user can save a workflow node that points to another user's private HTTP toolset by using a crafted saved tool id such as http-/. The normal toolset routes deny access, but the workflow...

CVE-2026-61643 FastGPT: workflow runtime can execute another user's private HTTP toolset

FastGPT is a knowledge-based AI application platform. From 4.14.17 until 4.15.0-beta5, an authenticated FastGPT user can save a workflow node that points to another user's private HTTP toolset by using a crafted saved tool id such as http-/. The normal toolset routes deny access, but the workflow...