kallurayaankit/jev-email-triage
User repository on GitHub
- Forks
- 0
- Last pushed
- Updated Oct 2026
Repo stats from the GitHub API, cached Oct 2026.
Jev directory
CommunityJev Agents & MCP
Jev Email Triage
A two-stage email triage agent built with Jev: stage one classifies each email, hard rules in code decide whether stage two runs, and stage two drafts a reply only when a reply is expected and it is safe to auto-send.
- Stars
- 0
- Language
- Python
- License
- MIT
- Reading time
- 2 min
- Published by
- Community
- Author
- kallurayaankit
- Last verified
- Oct 2026
- Languages
- Python
- Capabilities
- Open source
Highlights
- Two chained Jev calls: stage one classifies the email, stage two drafts a reply whose state includes stage one's output, so drafts know type and urgency.
- Hard rules run between the stages in code, not the model: spam and promotions never reach stage two, newsletters get muted, no-reply emails get archived.
- Quick start is uv sync plus copying .env.example to .env; Python 3.12, MIT license, CI badge.
- 0 stars, 0 forks; created and pushed 2026-10-12, found via a 'built with jev' README search.
Quickstart
bash
uv sync
cp .env.example .envWatch out
README reviewed 2026-10-17; the captured excerpt stops immediately after the quick-start commands, so the environment variables, Jev questions, run instructions and tests are not visible, and the repo has 0 stars.
Questions about this entry
Short answers drawn from the linked project or page.
A two-stage email triage agent built with Jev: stage one classifies the email, and stage two drafts a reply only if stage one says a reply is expected and only when it is safe to auto-send.
Stage two's state includes stage one's output, so the draft is written knowing the email is, for example, a work request, high urgency and from a colleague, as the README describes the architecture.
They run in code rather than the model: spam and promotions never reach stage two, newsletters get muted, and emails that do not need a reply get archived, all decided by rules.
The README quick start is uv sync followed by copying .env.example to .env, which suggests configuration or an API key in the environment; the captured excerpt cuts off right after those commands.
No. The README says stage two runs only if stage one says a reply is expected and only when it is safe to auto-send, so the second Jev call is gated by rules rather than always firing.
Python 3.12 with an MIT license and a CI workflow badge; GitHub metadata in the digest shows 0 stars and 0 forks, created and last pushed 2026-10-12.
The excerpt ends right after the .env copy step, so the Jev question definitions, environment variables, how to run the agent, tests and sample behavior are not documented in the digest.
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
A free Claude Code skill that puts Jev in front of the expensive model: thirteen commands sort, check, score, gate and verify with Jev, so Claude reads only the unsure slice and writes the output. Needs python3 and an OpenRouter key.
Jev Agents & MCP#community#python#claude-code
Community
charlesdove977/claude-x-jevPython
charlesdove977/claude-x-jevPythonDjaevin, a calibrated, local-first decision layer from Icaro0310 that runs behind a Devin ACP backend with no API key. It ships as an MCP server published in the official MCP Registry and installs from PyPI as poordjaevin; the repo topics name calibration, conformal prediction and NLI.
Jev Agents & MCP#community#github#python
Community
icaro0310/devin-judgePython
icaro0310/devin-judgePythonA harness in which a small decision model watches every step of your coding agent: routine steps go to a cheap model, hard steps to the strong one, aiming for the same results at a fraction of the bill. Python 3.10+, MIT, alpha, with an arXiv paper and Hugging Face models under the same name.
Jev Agents & MCP#community#github#python
A plugin-based agent harness (FastAPI + Vue 3) bundling a chat GUI, multi-provider model access, an MCP client and marketplace, skills, workflow studio, sandboxed execution and a Tauri desktop shell. It integrates TypeSafe AI's Jev for structured Choice, Score and Noul decisions.
Jev Agents & MCP#community#github#agents
Community
andyqiuqiubo/andy-harnessTypeScript
andyqiuqiubo/andy-harnessTypeScriptAn auto-mode safety monitor for the Pi coding agent: every tool call Pi is about to run gets one System One classifier question per rule, then allow, block or ask is decided in code. Built with DSPy against TypeSafe's Jev and exported to a Pi extension calling lunaroute/djev by default.
Jev Agents & MCP#community#github#pi
A local System One decision layer for AI coding agents, shipped as an MCP server and Agent Skill, that offloads high-volume classification, checks, scores and ranking from frontier LLMs to sub-50ms local models with calibrated output.
Jev Agents & MCP#community#github#mcp
Community
aktonay/snapdecPython
aktonay/snapdecPythonFrom 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.