Survival LLM
A free AI assistant and survival library (1.5 million offline articles) that runs on your own computer, with no internet. Shows the source of every answer. Mac and Windows.
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。
一分钟完成验证制作者的介绍仅有上面的一句话。 更多详情请阅读仓库 README。 仓库 README ↗
正在加载完整 README。
查看完整 README ↗Solartracker is the canonical repository home of the PV Manager rewrite.
PV Manager is a solar-energy and tariff-analysis web app focused on one core question:
Does owning solar materially reduce a household's electricity cost, and by enough to justify the install or financing cost over time?
The active product is the rewrite in apps/web.
It is the codebase validated for the main branch and the one all new product
work should target.
PV Manager combines provider telemetry, tariff history, and financial calculation logic to help a household understand:
The rewrite is intentionally built around trustworthy interpretation rather than just showing raw power numbers.
Live: current-day generation, import/export, and solar contextHistorical Day: detailed day view for a specific dateRange History: period-based analysis across weeks, months, years, and
custom rangesSettings: tariff management, provider setup, location, investment history,
and related user configurationAdmin / Approval flows: support for invited beta users and operator controlSome adjacent features are already scoped in the backlog but still in progress, including:
The current rewrite stack in apps/web is:
Next.js 15 with the App RouterReact 19TypeScriptTailwind CSSNextAuth for authenticationDrizzle ORM with PostgresVitest for testsECharts for historical and financial chartingSupporting integrations include:
MyEnergi for live and historical energy dataOpen-Meteo for current weather / forecast contextResend for email flowsAt a high level, the rewrite is split into:
app/: routes, pages, route handlers, and server-side entrypointssrc/domain/: business and financial calculation rulessrc/providers/: provider adapters and credential handlingsrc/jobs/: scheduled ingestion and summary generation logicsrc/db/: schema, queries, and seed datasrc/range/, src/live/, src/settings/, src/weather/: feature-focused
modulesThe important architectural idea is separation between:
That separation is what lets the product evolve financially without the UI being tightly coupled to raw provider payloads.
.
├── apps/
│ └── web/ # Active rewrite app
├── docs/ # Product brief, architecture, features, stories, decisions
└── output/ # Generated artifacts and local outputs
The rewrite app lives under apps/web.
cd apps/web
npm install
npm run dev
Useful commands:
npm run build
npm run test
npm run lint
npm run db:generate
npm run db:push
npm run db:seed
npm run db:studio
npm run job:daily-summary
npm run job:catch-up
The current example environment file is:
At minimum, local database access is expected. Additional auth, email, provider, and runtime secrets may be required depending on which flows you want to run.
The rewrite is moving toward a model where:
This matters because much of the product value is in accurate historical and financial interpretation, not just in displaying telemetry.
PV Manager is built with a doc-first delivery flow so product intent, calculation rules, and implementation scope stay aligned.
The typical path from idea to shipped feature is:
docs/features/.docs/decisions/.docs/stories/todo/.In practice, the repo is intended to answer these questions clearly:
What are we building? -> docs/product-brief.md, docs/use-cases.mdWhy does it matter? -> feature docs and decision recordsWhat exactly is in scope right now? -> docs/backlog.md and the active
story docsHow should it work? -> architecture and calculation docsWhat was actually delivered? -> code, tests, completed stories, and commit
historyThis approach is especially important here because the product is not just a UI project. Small wording, tariff, or data-model changes can alter the financial meaning of the app, so implementation is expected to stay tied closely to written product intent.
Key docs:
docs/product-brief.mddocs/architecture.mddocs/calculation-spec.mddocs/use-cases.mddocs/backlog.mdPlanning and execution docs:
docs/features/ for feature-level scopesdocs/stories/ for implementation slicesdocs/decisions/ for important architectural and product decisionsThe rewrite already includes meaningful end-to-end product slices, including:
The next major focus areas are improving Range History interpretation and performance, introducing estimate-aware handling for missing historical days, and continuing the move from raw interval-heavy request paths to better derived reporting layers.