Jev in production › Agent context and memory
Claude Code and Codex skill retrieves TypeSafe's Jev documentation into agent context so citations reference the correct page and heading; precision remains unmeasured.
Skills for Claude Code and Codex, and a study course, for building with Jev, TypeSafe's System One model. In the Ratings folder find the list of rated Jev projects and our eval methodology for user safety and quality.
Not every skill calls Jev: some are guides and LLM evaluators for building with it.
Name What it does Calls Jev Status ------------ Jevaluate Reads a Jev project's code and rates how well it uses Jev, with provenance for every finding No: an LLM rates, and code checks the rating Available Jevaluate, privately Rates a project you can't publish, such as a client's code or a local folder, with the same rubric, and keeps the rating private. Run /jevaluate-private No Available Jevaluate Harness Runs rating rounds, re-rates and rubric...
For the project's own README, linking back here: