DEMO / RECORDING
Overview & Result
An architecture connecting Claude to an MCP server and browser extension. Users converse naturally with Claude, while Jev evaluates browser DOM actions to fulfill goals without requiring users to micromanage web steps.
How Jev fits in the loop
- User tells Claude to perform an action on the web (e.g., find top posts on X).
- Claude issues tool requests via the AskJev MCP server.
- Jev scores and chooses the exact DOM interaction.
- The browser extension runs the interaction on the live tab.
How to reproduce
- Inspect the source implementation at https://github.com/ranjan2829/AskJev.
- Configure local Jev API credentials and register required tool/model bindings.
- Execute test prompts and observe decision latency and routing outputs.
Why this build matters
Exemplifies clean separation of concerns: generative models converse, Jev decides tactical actions.
Limitations
- Metrics are author-reported from the initial release unless independently verified.
- Requires access to the respective agent framework or runtime environment.
- Generative model execution remains external to the Jev decision step.