Continuous quality scores for code written by AI agents, from a local MCP plugin powered by Jev.
@niazmorshed_Runs on your machine as an MCP server and gives Claude Code, Codex, Cursor and OpenCode structured quality scores as they write code. The agent stays in charge of finding and fixing the weak spots; Jev just keeps score.
There is no hosted backend, database or tracking. Your API key stays on your machine, and the only request that leaves it goes straight to the Jev API.
An MCP server that puts Jev into the coding loop in Cursor, Codex and other agents.
Jev
TypeScript
MCPA single interface across dozens of AI providers, with a calibrated decide call powered by Jev for the routing and gating choices text generation was never meant to make.
JevTypeScript
Node.jsTrims long command output with Jev before Claude Code reads it, so later turns carry less noise.
JevTypeScript
Claude CodeRoute web requests by meaning instead of by path, in the Hono framework.
JevTypeScript
HonoTen ready-made Jev judgment tools for any agent: verify, screen, find, rerank, classify, decide and more.
Jev
JavaScriptMCPKeep the skills list out of every Pi request and let Jev pick the skills each task needs.
JevTypeScript
