Use after any web or API search, before opening results or spending another round. Jev picks which results to read, whether the evidence is enough, and which query to run next from ones you wrote.
Use to pick the cheapest model that is good enough for a turn — choosing a model, delegating a sub-task to a sub-agent, cutting model spend, or setting up and tuning Jev routing pools.
Use when unsure which of many installed skills applies to a request, if any, or when asked to make skill loading cheaper or more accurate. Jev ranks the whole catalog and may say no skill is needed.
Use when driving a desktop GUI through a computer-use driver — windows, menus, native apps, OS dialogs. You build a table of safe actions; Jev picks the next one in about 0.4 seconds.
Use on a mailbox export to sort mail into needs reply, updates, promotional, sales and spam — which messages are addressed to the person at all. For a support queue use jev triage.
Use when a transcript has to be cut to a fixed size and you must choose which turns go. Jev marks each turn keep, summarize or drop. Measured: it does not make a handoff better.
Use when Jev is not working yet, a Jev tool reports no_key or auth_failed, or the person asks to connect or fix Jev. Gets their TypeSafe or OpenRouter key into the secret store, unseen by you.
Use on passages a search just returned (memory, vault, session history, wiki, web) before reading them in. Jev ranks them, drops the irrelevant, and flags prompt injection hidden in the text.