Jev in production › Agent context and memory

JevTrace

Jev judges code relevance for an MCP server that trims coding-agent context to a token budget, based on the vendor's own 19-task retrieval benchmark (author).

Open on GitHub ↗

19measured, as published by the source
Use
Agent context and memory
Industry
Developer tools
Form
Open-source tool
Stage
Beta
Plugs into
MCP
Listed
2026-10-01
Found via
github
Repository
123wwwa/JevTrace
Stars
3
Forks
0
Last push
2026-10-01
Language
TypeScript
License
MIT

The README opens with

Retrieval benchmark on 19 JS/TS tasks with labels written by the JevTrace authors — see the method, per-task results and limitations .

Give it a coding task; it returns the JS/TS code that task needs, within a token budget. It uses the TypeScript compiler to follow calls, types, callers and tests, and uses the Jev decision model (through OpenRouter, TypeSafe, Vercel AI Gateway, OpenCode Zen or your own endpoint) to judge relevance.

Codex and Cursor take one config block each: see Claude Code, Codex and Cursor. To try it without an agent:

Badge

For the project's own README, linking back here:

Listed in Jev in production

Also used for agent context and memory