AI Decision System
Replayable, offline-verifiable records for AI-assisted go/no-go decisions: experiment promotions, typed model answers and human-approved actions.
翻译准备中
尚未检测到更新。GitHub 发布版本后会显示在这里。
个人 GitHub 账号下公开仓库的所有者可以登录并验证所有权,自行管理介绍,无需管理员批准。
登录并生成验证文件,在 30 分钟内将其提交到此仓库的默认分支。其他分支或分叉仓库中的文件无效。然后在下方检查文件,即可无需管理员批准直接管理项目。请保留文件以维持管理权限。
如需登录,将前往登录页面。登录后请回到此处,再次点击所选的验证按钮。
在项目文件夹中运行 /nomorevibe verify,即可验证所有权并自行管理介绍。
一分钟完成验证A small, modular Django project for basic server maintenance / monitoring workflows.
翻译准备中
取自仓库 README
正在加载完整 README。 以下为已保存的摘录。
A small, modular Django project for basic server maintenance / monitoring workflows.
This top-level README is the entry point and documentation hub. App-specific docs live alongside the apps.查看完整 README ↗A small, modular Django project for basic server maintenance / monitoring workflows.
This top-level README is the entry point and documentation hub. App-specific docs live alongside the apps.
docs/Architecture.mddocs/Installation.mddocs/Security.mdapps/checkers/README.mdapps/alerts/README.mdapps/notify/README.mdapps/intelligence/README.mdapps/orchestration/README.mdbin/README.mduv.lock)Dependencies (from pyproject.toml): Django + psutil.
See Installation document docs/Installation.md.
This project supports two modes — see Architecture for full details:
Quick examples:
# Pipeline mode (sync, with sample alert)
uv run python manage.py run_pipeline --sample
# Standalone health checks
uv run python manage.py check_health
# Run checks and generate alerts
uv run python manage.py run_pipeline --checks-only
This project supports dotenv files via python-dotenv.
.env by copying .env.sample..env.dev for dev-only defaults by setting DJANGO_ENV=dev.Common variables:
DJANGO_SECRET_KEY (required in production; local dev can fall back to an insecure default)DJANGO_DEBUG (1/0)DJANGO_ALLOWED_HOSTS (comma-separated)INBOX_DEPTH_WARN (drain backlog warning threshold; the pipeline is broker-free)This repo uses uv for dependency management and a small, consistent dev-tooling stack configured in pyproject.toml:
# Install runtime + dev tools
uv sync --extra dev
# Set up pre-commit hooks
uv run pre-commit install
# Run pre-commit manually on all files
uv run pre-commit run --all-files
# Format
uv run black .
# Lint (and auto-fix imports where possible)
uv run ruff check . --fix
# Tests
uv run pytest
# Test coverage (100% branch coverage required)
uv run coverage run -m pytest && uv run coverage report
# Optional: type-check
uv run mypy .
cp .env.sample .env
uv run python manage.py migrate
./bin/cli.sh
The CLI guides you through all available commands with their options.
./bin/install.sh aliases
After setup, use aliases like sm-check-health, sm-run-check, etc. See bin/README.md for the full alias list.
Alternatively, run commands directly:
# Run the health check suite
uv run python manage.py check_health
# List available checkers
uv run python manage.py check_health --list
(Optional) Run the Django server:
uv run python manage.py runserver
config/ — Django project settings/urls/asgi/wsgiapps/ — Django apps
apps/checkers/ — health checks + management commandsapps/alerts/ — alert ingestion (scaffold)apps/notify/ — notification drivers (scaffold)apps/intelligence/ — intelligence/recommendations systemapps/orchestration/ — pipeline orchestration (alerts → checkers → intelligence → notify)bin/ — shell scripts (installer, cron setup, interactive CLI)AGENTS.md describes the available agent roles and project rules.MIT License (c) 2026 Burak.
See LICENSE.