PYSEC-2026-7

Agenta is an open-source LLMOps platform. A Server-Side Template Injection SSTI vulnerability exists in versions prior to 0.86.8 in Agenta's API server evaluator template rendering. Although the vulnerable code lives in the SDK package, it is executed server-side within the API process when runni...

PYSEC-2026-6

Agenta is an open-source LLMOps platform. In Agenta-API prior to version 0.48.1, a Python sandbox escape vulnerability existed in Agenta's custom code evaluator. Agenta used RestrictedPython as a sandboxing mechanism for user-supplied evaluator code, but incorrectly whitelisted the numpy package ...

CVE-2026-27961

Agenta is an open-source LLMOps platform. A Server-Side Template Injection SSTI vulnerability exists in versions prior to 0.86.8 in Agenta's API server evaluator template rendering. Although the vulnerable code lives in the SDK package, it is executed server-side within the API process when runni...

CVE-2026-27961

Agenta (open-source LLMOps platform) has a Server-Side Template Injection (SSTI) vulnerability in API server evaluator templates for versions prior to 0.86.8. The vulnerable code runs server-side within the API process (SDK code executed server-side) and does not affect standalone SDK usage; impa...

EUVD-2026-8817

Agenta is an open-source LLMOps platform. A Server-Side Template Injection SSTI vulnerability exists in versions prior to 0.86.8 in Agenta's API server evaluator template rendering. Although the vulnerable code lives in the SDK package, it is executed server-side within the API process when runni...

CVE-2026-27961 Agenta's Server-Side Template Injection (SSTI) via custom evaluator Jinja2 templates allows RCE

Agenta is an open-source LLMOps platform. A Server-Side Template Injection SSTI vulnerability exists in versions prior to 0.86.8 in Agenta's API server evaluator template rendering. Although the vulnerable code lives in the SDK package, it is executed server-side within the API process when runni...

CVE-2026-27961 Agenta's Server-Side Template Injection (SSTI) via custom evaluator Jinja2 templates allows RCE

Agenta is an open-source LLMOps platform. A Server-Side Template Injection SSTI vulnerability exists in versions prior to 0.86.8 in Agenta's API server evaluator template rendering. Although the vulnerable code lives in the SDK package, it is executed server-side within the API process when runni...

CVE-2026-27952 Agenta has Python Sandbox Escape, Leading to Remote Code Execution (RCE)

Agenta is an open-source LLMOps platform. In Agenta-API prior to version 0.48.1, a Python sandbox escape vulnerability existed in Agenta's custom code evaluator. Agenta used RestrictedPython as a sandboxing mechanism for user-supplied evaluator code, but incorrectly whitelisted the numpy package ...

CVE-2026-27952

Summary of CVE-2026-27952 (Agenta) : The vulnerability affects the Agenta-API (self-hosted API server) prior to version 0.48.1. A Python sandbox escape in the custom code evaluator used RestrictedPython, but the sandbox allowlist erroneously included the numpy package. This allowed authenticated ...

EUVD-2026-8814

Agenta is an open-source LLMOps platform. In Agenta-API prior to version 0.48.1, a Python sandbox escape vulnerability existed in Agenta's custom code evaluator. Agenta used RestrictedPython as a sandboxing mechanism for user-supplied evaluator code, but incorrectly whitelisted the numpy package ...

PT-2026-22222

Name of the Vulnerable Software and Affected Versions Initiative versions prior to 0.32.4 Description Initiative is a self-hosted project management platform vulnerable to Stored Cross-Site Scripting XSS in the document upload functionality. Users with upload permissions within the "Initiatives"...

PT-2026-22210

Name of the Vulnerable Software and Affected Versions Hoppscotch versions prior to 2026.2.0 Description Hoppscotch, an API development ecosystem, had a critical security issue where an unauthenticated attacker could overwrite the entire infrastructure configuration of a self-hosted instance. This...

wger security vulnerabilities

WGER is an open-source project developed by the WGER Team, written in Django, and serves as a self-hosted FLOSS fitness/exercise, nutrition, and weight tracking application. Versions of WGER 2.4 and earlier contained security vulnerabilities, which were caused by improper handling of cache key...

PT-2026-22223

Name of the Vulnerable Software and Affected Versions Initiative versions prior to 0.32.4 Description Initiative, a self-hosted project management platform, does not invalidate previously issued JWT access tokens after a user changes their password. This allows older tokens to remain valid until...

PT-2026-22213

Name of the Vulnerable Software and Affected Versions Manyfold versions prior to 0.133.1 Description Manyfold is a self-hosted web application for managing 3d models. A flaw exists in the get model method within the ModelFilesController lines 158-160 where models are loaded using Model.find...

PT-2026-22180

Name of the Vulnerable Software and Affected Versions hexpm versions prior to 5d2ccd2f14f45a63225a73fb5b1c937baf36fdc0 Description A path traversal issue exists in hexpm’s Local Storage backend, impacting self-hosted deployments. The issue resides within the 'Elixir.Hexpm.Store.Local' module and...

wger security vulnerabilities

WGER is an open-source project developed by the WGER Team, written in Django, and serves as a self-hosted FLOSS fitness/exercise, nutrition, and weight tracking application. Versions of WGER 2.4 and earlier contained security vulnerabilities; these vulnerabilities stemmed from bypassing user-scop...

Manyfold security vulnerabilities

Manyfold is a self-hosted web application developed by Manyfold OpenSource. Versions of Manyfold prior to 0.133.1 contained a security vulnerability, which was caused by the getmodel method in the ModelFilesController bypassing Pundit authorization...

CVE-2026-27116

Vikunja is an open-source self-hosted task management platform. Prior to version 2.0.0, a reflected HTML injection vulnerability exists in the Projects module where the filter URL parameter is rendered into the DOM without output encoding when the user clicks "Filter." While and are blocked, , ,...

EUVD-2026-8719

ServiceNow has addressed a remote code execution vulnerability that was identified in the ServiceNow AI platform. This vulnerability could enable an unauthenticated user, in certain circumstances, to execute code within the ServiceNow Sandbox. ServiceNow addressed this vulnerability by deploying...