User repository on GitHub
Turn an inbox into a short action queue with Jev (TypeSafe System One)
- Forks
- 0
- Last pushed
- Updated Sep 2026
Repo stats from the GitHub API, cached Oct 2026.
Jev directory
CommunityJev Tools & CLIs
Jev Inbox Queue
Turns an inbox into a short action queue: Jev answers seven typed questions per email thread in one request, plain Python rules decide what enters the queue, and cached answers mean changing a rule never needs new API calls.
- Stars
- 1
- Language
- Python
- License
- MIT
- Reading time
- 3 min
- Published by
- Community
- Author
- tusharck
- Last verified
- Oct 2026
Highlights
- Seven questions per thread in one request: noul for needs-action, someone-waiting and promotional; choice for kind and next step; score for urgency.
- Demo inbox: 47 hand-labelled emails, 0 of 23 actionable missed, 0 of 23 irrelevant shown, promotional question correct 47 of 47, all sorted in 1.4 s.
- Real inbox: 100 emails sorted in 2.7 seconds — 5 need action, 1 to check, 94 filtered out, 74 of them promotional.
- Policy stays in code: needs_action above 0.6 enters the queue and is_promotional above 0.5 is filtered; cached answers make threshold changes free.
- The Gmail path reads All Mail via an app password stored in the system keychain, not in a file, so sent replies are visible to Jev.
Quickstart
bash
uv sync
cp .env.example .env
uv run python -m inbox_queue demo
uv run python -m inbox_queue evaluateWatch out
Copy reviewed against the repo README on 2026-10-15. The author is explicit that the demo numbers come from emails he wrote and labelled himself, that he adjusted the promotional filter after seeing results, and that your own inbox is the real test; no independent benchmarks exist.
Questions about this entry
Short answers drawn from the linked project or page.
It reads each email thread once and shows only the ones that need you, turning the inbox into a short action queue with what needs you, how urgent it is and the next step, instead of making you re-read everything to find the few emails that matter.
Jev answers seven typed questions per thread in one request, in parallel: needs-action, someone-waiting and promotional as noul; kind and next step as choice; urgency as score. Jev makes the judgments — the queue rules stay in plain Python in policy.py.
No, and the README says so directly: the author wrote and hand-labelled the 47 demo emails, adjusted one rule after looking at the results, and calls your own inbox the real test. Treat the numbers as a smoke test, not a benchmark.
Clone the repo, run uv sync, copy .env.example to .env and add your TYPESAFE_API_KEY, then run uv run python -m inbox_queue demo to open the queue for the fake inbox, or evaluate for the missed and irrelevant counts.
Yes. Enable 2-Step Verification and create an app password, set IMAP_USER and MY_NAME in .env, save the app password to the system keychain with uv run python -m inbox_queue save-password, then start with imap --days 3 --limit 10.
No. Answers are cached, so changing a rule never needs new API calls; for example uv run python -m inbox_queue evaluate --high 0.5 --low 0.2 re-scores the cached answers. Only fresh email threads go back to Jev.
It reads All Mail, which includes your sent replies, so Jev can see what you have already answered. That is deliberate: it is how the queue knows whether a person is still waiting on you. Keep that in mind before pointing it at a work account.
Reactions & coverage
Posts, threads, and videos about this entry from around the web.
X: 1,565 emails, six decision models
X: Fifty emails in under two seconds
More Jev resources like this
Toolkit, knowledge base and local dashboard for building with Jev via OpenRouter Alpha Decisions or the TypeSafe Direct API: a zero-dependency SQLite request log, a localhost REST server and Python CLI demos.
Jev Tools & CLIs#community#catalog-import#logicrw
Community
rileycarney/jevtoolsHTML
rileycarney/jevtoolsHTMLA small desktop console and CLI for Jev: paste a state, add noul, choice and score questions by hand, and read calibrated probabilities back without writing code. Standard library only, with a tkinter window whose labels are in Chinese.
Jev Tools & CLIs#community#catalog-import#logicrw
Community
wenchenxi/jev-consolePython
wenchenxi/jev-consolePythonA local, open-source personal AI assistant for your job: mail, chats and tickets become tasks, agents you already use such as Claude Code, Codex and Gemini do the work, and nothing is sent until you approve. The current release is v0.3.8.3.
Jev Tools & CLIs#community#catalog-import#logicrw
Community
ldbumble/taskuaryPython
ldbumble/taskuaryPythonA Rust CLI that streams Kubernetes logs, triages them with local offline rules or typed Jev questions online, and presents an interactive terminal UI plus JSON reports. Collection is read-only and every judgment is advisory.
Jev Tools & CLIs#community#catalog-import#logicrw
Community
sunil-sadasivan/jevernetesGo
sunil-sadasivan/jevernetesGoAn Ollama-style local runner for System One decision models from Hugging Face: pull, serve and query them behind the same /v1/systemone API as TypeSafe's Jev, with a terminal model manager and a browser playground.
Jev Tools & CLIs#community#catalog-import#madewithjev
Community
nvkudva/ollajevPython
nvkudva/ollajevPythonA Windows fork of the jev-chat WeChat assistant that fixes DPI-unaware screen capture on scaled displays and adds an optional key-free local laya judge — with the author's own measurements showing that local engine is not yet usable for Chinese chat.
Jev Tools & CLIs#community#catalog-import#logicrw
From the Jev guides
Original write-ups on what this entry is for.
From the Jev AI model community
Posts from builders shipping with Jev right now.