lunary-ai/lunary version 1.0.1 is vulnerable to improper authorization, allowing removed members to read, create, modify, and delete prompt templates using an old authorization token. Despite being removed from an organization, these members can still perform operations on prompt templates by sending HTTP requests with their previously captured authorization token. This issue exposes organizations to unauthorized access and manipulation of sensitive template data.
The product performs an authorization check when an actor attempts to access a resource or perform an action, but it does not correctly perform the check.
Link | Tags |
---|---|
https://huntr.com/bounties/671bd040-1cc5-4227-8182-5904e9c5ed3b | patch exploit third party advisory issue tracking |
https://github.com/lunary-ai/lunary/commit/d8e2e73efd53ab4e92cf47bbf4b639a9f08853d2 | patch |