CVE-2026-121936 · Package
NO FIX YET · SEE WHAT TO DO
easyVoice before 0.0.23 lets an unauthenticated caller supply an openaiBaseUrl when useLLM is enabled, making the server send POST requests to attacker-chosen addresses. The reflected error text in the x-generate-tts-warning header lets an attacker probe loopback and internal hosts and ports.
first public Nov 4, 2026 · 1 source
Is this in your stack? Find out in 2 minutes.
Tell InfraStack what you run. It checks this one, and every next one, against your stack.
Ask from here. InfraStack checks the affected versions against your stack and answers in your inbox.
InfraStack checks the affected versions against your stack and answers in Slack, with the evidence and the fix.
Teams running easyVoice as a server that accepts TTS requests with useLLM enabled are exposed, since the openaiBaseUrl value comes from the request. Deployments that keep useLLM off, or that run easyVoice only as a local library without a network-facing endpoint, are less exposed. Internal services reachable from the server host are the realistic targets.
Not affected if default-config, no-network-endpoint
Who cares easyvoice, node, self-hosted
npm ls easyvoice or the package.json dependency.useLLM enabled and an openaiBaseUrl accepted from request input on the /api/v1/tts endpoints.0.0.23 or later.easyVoice before 0.0.23 contains a server-side request forgery vulnerability that allows unauthenticated attackers to make the server send POST requests by supplying openaiBaseUrl with useLLM enabled. Attackers can target loopback and internal services through /api/v1/tts endpoints, using reflected error messages in the x-generate-tts-warning header to probe internal hosts and ports.
This page as .md · .json · in your agent: get_signal("CVE-2026-121936") · CVSS:4.0/AV:N/AC:L/AT:N/PR:N/UI:N/VC:L/VI:L/VA:N/SC:N/SI:N/SA:N/E:X/CR:X/IR:X/AR:X/MAV:X/MAC:X/MAT:X/MPR:X/MUI:X/MVC:X/MVI:X/MVA:X/MSC:X/MSI:X/MSA:X/S:X/AU:X/R:X/V:X/RE:X/U:X
NVD data is provided by NIST and is public domain. Posts quoted belong to their authors.