It probably won't be secure. Cryptography and security are unforgiving disciplines that are rife with subtle pitfalls. You don't get points for a pretty good solution. Unless you know for sure that you can (rigorously) cover all the bases, you'll be leaving a door wide open. Half-baked secrets management systems are a go-to target for attackers.
It probably won't be reliable. Configuration and secrets management systems need to serve a large number of requests, and can easily become single-points-of-failure for critical services if they aren't achitected properly. Consistency is also essential: config values that aren't in sync can cause bugs, production outages, and other forms of chaos.
It probably won't be easy to work with. How much time are you really going to spend polishing the UX/DX of your configuration and secrets management system? Let's be honest: while it might fit into your workflow, it's also going to have plenty of warts, a barebones feature-set, not-enough-documentation, a steep learning curve for new devs, and a headache-quotient that grows over time as you integrate it with new apps and services.
It will probably take a long time to build. If you aren't convinced of that yet, I'd suggest re-reading the previous few bullet points again. Can you prove me wrong on every point and build a highly secure, highly reliable, strongly consistent, easy-to-use-and-integrate configuration and secrets management system? Maybe you can. Maybe. But one thing is certain: it won't be quick.