Jev agent folder: Claude only writes, Jev decides, code acts
A folder structure where Jev handles routing, worker selection, relevance, completion checks and risk judgment, code performs actions and leaves decision receipts, with 10,000 decisions costing $0.42 at a 300ms median.
I STOPPED LETTING CLAUDE MAKE DECISIONS THE DAY I BUILT THIS JEV AGENT FOLDER, AND IT'S STRAIGHT F*CKING GOLD I used to let Claude decide, write and act on every single step -> now Claude only writes. Jev makes the calls, code does the acting, and every step leaves a receipt what lives in the folder: • the input > AGENTS.md - when to call Jev and when to skip it > state/build_state.py - goal, workers, done, missing, constraint. raw evidence, never a summary • Jev decides (questions/) > http://route.py - which model tier takes the task > next_worker.py - which worker moves next > http://relevance.py - keep or drop each tool output > http://done.py - is the goal really met > http://risky.py - will this send, pay or delete? • code acts (rules/) > hard_rules.py - stop after ten actions, never publish anything unapproved > thresholds.yaml - act only on confident answers. fraud needs 0.95 • Claude writes (workers/) > http://research.py - sources and notes > http://writer.py - drafts and briefings > the one place in the folder where text gets generated • the proof > receipts/decisions.jsonl - options offered, chosen id, re-check, fallback > evals/ - dozens of my own labelled traces. thresholds tuned on them, confirmed on held-out • the guards (hooks/) > pre_tool_use.sh - every command checked before it runs > http://stop.sh - verifies "all done" before the agent can stop 5 questions, one call per turn 10,000 decisions for $0.42, median 300 ms an LLM writes, Jev decides, code acts
This post quotes another post: x.com/mr_buzzoni/status/2103689373774483815