Hello all,
Thank you all for the vocal and clear feedback. We intended this change to simplify the setup for developers who use agents to build APIs with coding agents as we see many people struggling in wiring these up. Installing skills for coding agents is a new way of working with Postman and we thought that the in-product and website messaging conveys the intent.
Based on the feedback from the community, we understand this wasn’t the case so we will be re-designing the experience with the right level of control.
In the meantime, we are reverting this change.
Thinking about that, it could be neat for a threat actor to have their hands on one's coding agents' config, so whatever it puts out, it leaves predefined vulnerabilities or sends parts of the closed source code back home.
It'd double scarry if this 'update' could reconfig not only outsourced models like Claude, but local ones too.
1 comment