惯性聚合 高效追踪和阅读你感兴趣的博客、新闻、科技资讯
阅读原文 在惯性聚合中打开

推荐订阅源

Hugging Face - Blog
Hugging Face - Blog
Recent Announcements
Recent Announcements
V
Visual Studio Blog
博客园 - 叶小钗
H
Help Net Security
aimingoo的专栏
aimingoo的专栏
宝玉的分享
宝玉的分享
U
Unit 42
钛媒体:引领未来商业与生活新知
钛媒体:引领未来商业与生活新知
F
Fortinet All Blogs
V
V2EX
Stack Overflow Blog
Stack Overflow Blog
WordPress大学
WordPress大学
D
DataBreaches.Net
J
Java Code Geeks
H
Hackread – Cybersecurity News, Data Breaches, AI and More
A
About on SuperTechFans
酷 壳 – CoolShell
酷 壳 – CoolShell
量子位
C
Check Point Blog
奇客Solidot–传递最新科技情报
奇客Solidot–传递最新科技情报
小众软件
小众软件
Microsoft Azure Blog
Microsoft Azure Blog
M
MIT News - Artificial intelligence

Hacker News - Newest: "AI"

AI can't read an investor deck AI as an attorney? Student uses ChatGPT, Gemini to sue UW over alleged racial discrimination Hacking MCP Servers in AI Systems – The Rug Pull: Tool Changes After Approval GitHub - MeepCastana/KubeezCut: Free Web based video editor GitHub - GenAI-Gurus/awesome-eu-ai-act: Curated tools, official sources, OSS, templates, and guides for EU AI Act compliance. Can AI judge journalism? A Thiel-backed startup says yes, even if it risks chilling whistleblowers Coming soon: 10 Things That Matter in AI Right Now DARPA built an AI to fact-check enemy weapons claims What explains heterogeneity in AI adoption? When AI Meets Muscle: Context-Aware Electrical Stimulation Promises a New Way to Guide Human Movements - Department of Computer Science AI Changed How We Build. It Did Not Change What Matters. Linux rules on using AI-generated code - Copilot is OK, but humans must take 'full responsibility for the… Meta spins up AI version of Mark Zuckerberg to engage with employees Code Mode: Let Your AI Write Programs, Not Just Call Tools | TanStack Blog GitHub - Delavalom/graft: Go framework for building AI agents. Type-safe tools, multi-provider (OpenAI, Anthropic, Gemini, Bedrock), zero vendor SDKs. India's TCS tops estimates, says new AI models did not dent services demand Gen Z's fading AI hype Strong feeling: we are in a folded AI reality GitHub - machinarii/total-recall-catalog: A reference catalog of latest knowledge retrieval, memory & RAG systems GitHub - mensfeld/code-on-incus: Give each AI agent its own isolated machine with root, Docker, and systemd. Active defense detects and stops threats automatically.. Quantization, LoRA, and the 8% Problem: Benchmarking Local LLMs for Production AI Iran war: We spoke to the man making Lego-style AI videos that experts say are powerful propaganda Powell, Bessent discussed Anthropic's Mythos AI cyber threat with major U.S. banks GitHub - immartian/bellamem: Persistent belief-graph memory for AI agents. Retrieves decisive context by importance — not recency, not RAG, not /compact. recursive-mode: The Repo-Native Operating System for AI Engineering After the attack on Sam Altman's home, will AI CEO's go on the offensive? The biggest advance in AI since the LLM Opus 4.6 vs GPT 5.4 One Prompt Unity World Generation Test “AI polls” are fake polls Client Challenge
GitHub - HimashaHerath/github-dev-wrapped: AI-powered wee...
2026-04-11 · via Hacker News - Newest: "AI"

npm

AI-powered weekly and monthly GitHub activity reports — deployed automatically to GitHub Pages, with zero backend infrastructure.

Like Spotify Wrapped, but for your code. Runs on your own GitHub account every week or month, generates a unique AI-written personality profile from your actual activity, and publishes a beautiful static report to GitHub Pages.

Live demo → · npm package →


What you get

Every run produces:

  • A full static HTML report page (no JavaScript, SEO optimized)
  • An animated SVG card to embed in your GitHub profile README

Two themes:

  • noir — dark, cinematic, monospace. Movie poster feel.
  • clean — light, minimal, editorial. Portfolio-ready.

Two frequencies:

  • weekly — runs every Monday, reports the previous complete week (Mon–Sun)
  • monthly — first of each month, richer stats

AI writes three things every run:

  1. Archetype — a unique sentence describing your coding personality this period, generated fresh from actual behavior signals. Never a fixed label.
  2. Evolution story — one paragraph comparing this period to your history. Unlocks after 2+ periods.
  3. Headline — one punchy line for the animated SVG card.

Quick start

npx github-dev-wrapped setup

The wizard takes about 60 seconds and handles everything:

  • Creates your report repo
  • Stores your API keys as encrypted GitHub secrets
  • Enables GitHub Pages
  • Commits the workflow file
  • Triggers your first report

You need:

  • A GitHub Personal Access Token with repo and workflow scopes
  • An API key from any supported LLM provider (free tiers work)

Embed the card in your profile README

After your first run, add this to your GitHub profile README.md:

<!-- Dark card -->
<img src="https://{username}.github.io/{repo}/screenshots/card-dark.svg" alt="Dev Wrapped" />

<!-- Light card -->
<img src="https://{username}.github.io/{repo}/screenshots/card.svg" alt="Dev Wrapped" />

Supported LLM providers

All free tiers on OpenRouter, Groq, and Gemini make the full feature set available at $0/month.

Provider Free tier Default model
OpenRouter Yes (many free models) meta-llama/llama-3.3-70b-instruct:free
Groq Yes llama-3.3-70b-versatile
Gemini Yes gemini-2.0-flash
OpenAI No gpt-4o-mini
Anthropic No claude-haiku-4-5-20251001
Grok (xAI) No grok-3-mini

What gets tracked

Activity stats (from GitHub API):

  • Commits — time of day, day of week, repo
  • PRs opened / merged / reviewed
  • Issues opened / closed
  • Languages written (by primary repo language)

Derived signals:

  • Peak coding hour
  • Coding streak (consecutive active days)
  • Focus score — how concentrated your effort was (one repo vs many)
  • Review generosity — helping others vs building your own stuff
  • Language drift — stack shifting over time

Evolution delta: Every run computes a delta against the previous period. Stored as JSON in your report repo so it compounds over time. Evolution data appears in reports after 2+ periods. The full narrative arc unlocks after 3+ months.


How it works

GitHub Actions (weekly/monthly cron)
    │
    ├─ fetch      Collect activity from GitHub API → write JSON to repo
    ├─ generate   Send signals to LLM → get archetype + headline + evolution
    ├─ render     Build HTML report + animated SVG card from Handlebars templates
    └─ deploy     Force-push static output to gh-pages branch

All data lives in your report repo as JSON. No external database. No hosted backend. Period history is committed to the main branch under data/; gh-pages serves the rendered output.


Configuration

The generated workflow file (.github/workflows/wrapped.yml) supports these inputs:

Input Default Description
frequency weekly weekly or monthly
theme noir noir or clean
language en Report language (en, ja, zh-CN, zh-TW, ko, es, fr, de, pt, ru)
timezone UTC IANA timezone (e.g. Asia/Colombo)
llm-provider Provider name (see table above)
llm-model Model name
date Target date YYYY-MM-DD for manual backfills via workflow_dispatch. Omit for automatic behaviour (weekly defaults to the previous complete week, monthly to the current month).

Cost

$0/month on a public repo.

  • GitHub Actions: free for public repos
  • LLM: one API call per period — OpenRouter/Groq/Gemini free tiers cover it
  • GitHub Pages: free for public repos

CLI commands

The npm package exposes internal commands used by the GitHub Action:

npx github-dev-wrapped setup          # Interactive setup wizard
npx github-dev-wrapped fetch          # Collect GitHub activity
npx github-dev-wrapped generate       # Generate AI narrative
npx github-dev-wrapped render         # Build HTML report + SVG card
npx github-dev-wrapped deploy         # Push to GitHub Pages

Contributing

Contributions are welcome — bug fixes, new themes, additional LLM providers, language translations, or anything that makes the reports better.

Getting started:

git clone https://github.com/himasha/github-dev-wrapped
cd github-dev-wrapped
npm install
npm run build       # compile TypeScript
npm test            # run tests

Key areas to contribute:

  • src/collector/ — GitHub API queries and signal derivation
  • src/llm/providers/ — add a new LLM provider (implement the LLMProvider interface)
  • src/renderer/themes/ — add a new visual theme (Handlebars templates + styles)
  • src/i18n/locales/ — add or improve a language translation

Please open an issue before starting larger changes so we can align on approach. PRs should include tests where applicable and pass npm test cleanly.

Releasing a new version:

npm version patch   # or minor / major
git push --follow-tags

That's it — pushing the tag triggers the publish workflow which runs tests, builds, publishes to npm, and creates a GitHub Release automatically.


License

MIT