I have been working on AI agent Flows for some time now and always found it annoying to have to redeploy my application for small changes to flows, prompts, model definitions etc.
This is why I build Flowshapr (https://flowshapr.ai). It let's you create and manage AI flows and prompts using a simple UI and then execute those flows with a small client side SDK and a one line code integration.
The tool is based on Genkit and will generate Genkit code on the fly if you want to deploy yourself.
This early version can manage simple flows, prompts, generate Genkit code, execute flows remotely and show traces. I am currently working on vector store integrations and more tools.
I have been working on AI agent Flows for some time now and always found it annoying to have to redeploy my application for small changes to flows, prompts, model definitions etc.
This is why I build Flowshapr (https://flowshapr.ai). It let's you create and manage AI flows and prompts using a simple UI and then execute those flows with a small client side SDK and a one line code integration.
The tool is based on Genkit and will generate Genkit code on the fly if you want to deploy yourself.
This early version can manage simple flows, prompts, generate Genkit code, execute flows remotely and show traces. I am currently working on vector store integrations and more tools.
The repo is here: https://github.com/flowshapr/flowshapr
I'd love to hear your feedback and learn more about your use cases!