メニュー
Optimize your GitHub profile as a living portfolio — craft a standout profile README, pin the right repos, tag for discoverability, and signal real builder credibility to hiring managers evaluating AI-native developer candidates.
Codex または Claude でインストール この Prompt をコピーして Codex、Claude、または他のアシスタントに貼り付けると、Skill ページを確認してインストールできます。
SOC 職業分類に基づく
| name | github-profile |
| description | Optimize your GitHub profile as a living portfolio — craft a standout profile README, pin the right repos, tag for discoverability, and signal real builder credibility to hiring managers evaluating AI-native developer candidates. |
Your GitHub profile is your portfolio, your proof of work, and often the first technical impression a hiring manager gets. This skill turns it into a strategic asset — not a graveyard of half-finished tutorials.
"Show what you build, not what you star." Hiring managers spend 30 seconds on your profile. Every element — README, pinned repos, bio, activity graph — should answer one question: Can this person ship?
GitHub renders a special README.md as your profile page if you create a repo that matches your username.
1. Create a new repo named exactly your GitHub username
→ github.com/yourname/yourname
2. Check "Add a README file" during creation
3. The README.md in this repo becomes your profile landing page
If the repo already exists but is private, make it public. GitHub only renders public profile READMEs.
Write it in this order. Every section earns its place or gets cut.
# Headline (1 line)
What you do + what makes you different. Not your job title — your value proposition.
## About
2-3 sentences. What you're building, what you care about technically,
what you're looking for. Written for a human, not a search engine.
## Tech Stack
Grouped by category. Only include tools you can talk about in an interview.
## Featured Projects
3-4 repos with one-line descriptions and what makes each interesting.
Link directly to the repos.
## Contact
LinkedIn, email, portfolio site. Make it easy to reach you.
| Weak | Strong |
|---|---|
| "Hi, I'm Alex" | "Building AI-powered developer tools in Python and TypeScript" |
| "Full-stack developer" | "Staff engineer turning LLM prototypes into production systems" |
| "Passionate about code" | "Shipping AI agents that automate real workflows" |
The headline should pass the "So what?" test. If a hiring manager reads it and thinks "so what?" — rewrite it.
Write 2-3 sentences that answer:
## About
I build AI-native applications that go beyond chat wrappers — tool-using agents,
retrieval-augmented generation pipelines, and LLM evaluation frameworks. Currently
focused on making AI systems reliable enough for production workloads. Open to
senior/staff roles where I can ship AI products end-to-end.
Do not write a life story. Do not list soft skills. Do not say "passionate learner."
Group by function. Only list what you could whiteboard in an interview.
## Tech Stack
**AI/ML:** Python, LangChain, OpenAI API, RAG pipelines, vector databases (Pinecone, pgvector)
**Backend:** Node.js, FastAPI, PostgreSQL, Redis
**Infrastructure:** AWS (Lambda, ECS, S3), Docker, Terraform
**Frontend:** React, TypeScript, Next.js
| Do | Don't |
|---|---|
| Group logically (AI/ML, Backend, Infra, Frontend) | Dump 40 badges in a grid |
| Include only tools you've used in projects | List every technology you've seen a tutorial for |
| Put your strongest/most relevant stack first | Alphabetize everything |
Skip the shields.io badge wall. A clean text list signals confidence. A wall of badges signals insecurity.
Link 3-4 of your best repos. Each gets one line explaining what it does and one line explaining why it's interesting technically.
## Featured Projects
**[ai-code-reviewer](https://github.com/yourname/ai-code-reviewer)**
Automated PR reviewer using GPT-4 with AST-aware context windowing. Reduced review time by 40% on a 5-person team.
**[rag-eval-framework](https://github.com/yourname/rag-eval-framework)**
End-to-end evaluation pipeline for RAG systems — measures retrieval precision, answer faithfulness, and hallucination rate.
**[deploy-cli](https://github.com/yourname/deploy-cli)**
Zero-config deployment tool for containerized Python apps. One command from code to running on AWS ECS.
Every project description should include a concrete result or technical detail, not just what the project "is."
## Get in Touch
- LinkedIn: [linkedin.com/in/yourname](https://linkedin.com/in/yourname)
- Email: you@example.com
- Portfolio: [yoursite.dev](https://yoursite.dev)
Make it frictionless. If someone wants to hire you, don't make them hunt for a way to reach you.
GitHub lets you pin up to 6 repos. These are the only repos most people will ever look at.
Pin repos that show:
1. You can ship a complete project (not just a starter template)
2. You understand the domain hiring managers care about (AI, infra, full-stack)
3. You write code that other humans can read
4. You solve real problems, not toy examples
| Pin This | Skip This |
|---|---|
| Original project with a real README | Fork of a popular repo with no changes |
| Tool/library that solves a specific problem | "awesome-list" or curated link collections |
| Project with tests, docs, and clean commits | Tutorial follow-along (todo app, calculator) |
| Collaborative project where you're a key contributor | Repo with 1 commit from 2 years ago |
| Something you can demo and explain in 5 minutes | Class homework or bootcamp assignments |
GitHub displays pinned repos left-to-right, top-to-bottom. Put your strongest project in position 1 (top-left). Most people only look at the first 2-3.
Your 6 pins should collectively answer: "What kind of engineer is this person?"
Example narrative: "AI-native full-stack developer"
Pin 1: AI agent framework (shows AI depth)
Pin 2: Production API with AI integration (shows backend + AI)
Pin 3: CLI tool for developer workflows (shows shipping real tools)
Pin 4: Open source contribution to known project (shows collaboration)
Pin 5: Infrastructure-as-code repo (shows production awareness)
Pin 6: Side project with a live demo (shows end-to-end ability)
Every pinned repo needs three things before it's profile-ready.
The short description shown on your profile and in search results. Write it like a commit message — clear, specific, no filler.
Bad: "A cool project I built"
Good: "CLI tool that generates type-safe API clients from OpenAPI specs"
Bad: "Machine learning stuff"
Good: "RAG evaluation framework measuring retrieval precision and answer faithfulness"
Topics make your repos discoverable in GitHub search and signal what technologies you work with.
Add 5-10 relevant topics per repo:
ai, llm, rag, python, fastapi, openai, langchain,
vector-database, developer-tools, cli
| Do | Don't |
|---|---|
| Use specific, searchable terms | Add generic tags like "code" or "project" |
| Include the primary language | Tag languages you didn't actually use |
| Add the problem domain (e.g., "developer-tools", "devops") | Add 30 tags hoping for SEO |
| Match terms hiring managers search for | Make up tags nobody searches for |
Every pinned repo needs a README that covers:
- What it does (1-2 sentences)
- Why it exists / what problem it solves
- How to install and run it
- Architecture or design decisions (brief)
- Screenshots or demo GIF if applicable
A repo without a README is a repo nobody will look at. A README without install instructions is a project nobody will try.
Use a clear, professional headshot. Not a logo, not an avatar, not a photo from 2014. Hiring managers are evaluating whether they want to work with you — a real face builds trust.
This appears on your profile and in search results. Treat it like a tweet.
Bad: "I like coding and learning new things"
Good: "Building AI developer tools. Python, TypeScript, LLMs. Open to staff-level roles."
Bad: "Student | Dreamer | Coder"
Good: "ML engineer shipping RAG pipelines to production. Previously @CompanyName."
Include: what you do, your primary stack, and your current status (open to work, building X, etc.).
GitHub gives you fields for:
Fill every field. Empty fields look like you don't care.
The green contribution graph is the first thing many recruiters look at. Here's what they actually read from it.
| Pattern | Signal |
|---|---|
| Consistent daily/weekly activity | Active builder, codes regularly |
| Large gaps with sudden bursts | Only codes for job searches (red flag) |
| Heavy activity on weekdays only | Mostly work contributions (fine, but not standout) |
| Steady mix of weekday + weekend | Genuine interest beyond the job |
| All green, every single day | Probably gaming the graph (also a red flag) |
The graph reflects:
- Commits to default branches
- Opening issues and PRs
- Creating repos
- Reviewing PRs
It does NOT reflect:
- Commits to non-default branches (until merged)
- Private repo activity (unless you enable it in settings)
- Starring or forking repos
Enable private contributions: Settings > Profile > check "Include private contributions on my profile." This fills in gaps from work activity without exposing code.
Do not manufacture commits to make the graph green. Hiring managers who care about AI-native developers care about what you build, not how green your calendar is. A sparse graph with 3 excellent pinned repos beats a solid green wall with nothing worth clicking on.
| Anti-Pattern | Why It Hurts You |
|---|---|
| Profile README with animated GIFs, visitor counters, and Spotify widgets | Looks like MySpace, not a professional portfolio |
| 30+ pinned badge icons for every framework ever | Signals breadth without depth — "jack of all trades" |
| Pinning forks you didn't meaningfully contribute to | Hiring managers check commit history. They'll notice. |
| "Hello world" or tutorial repos visible on profile | Suggests you're still learning the basics |
| No README on any repo | "I write code but can't communicate about it" |
| Bio says "aspiring developer" or "learning to code" | Undercuts your credibility before they look at anything |
| Dozens of repos with 1 commit each | Starts things, finishes nothing |
| Star count as a personality trait | Nobody cares how many repos you've starred |
| README that's 500 lines with every GitHub stat widget | Overcompensating. Let the work speak. |
| Committing API keys, .env files, or secrets | Instant disqualification. Audit your repos now. |
Run through this before sharing your profile with anyone.
Profile Basics:
[ ] Professional headshot as profile photo
[ ] Bio filled with role + stack + status (160 chars)
[ ] Website field links to portfolio or LinkedIn
[ ] Location set (city or timezone)
[ ] Company field filled (employer or "Open to opportunities")
[ ] Social accounts linked
Profile README (username/username repo):
[ ] Repo exists, is public, contains README.md
[ ] Headline passes the "So what?" test
[ ] About section is 2-3 sentences, not a novel
[ ] Tech stack grouped by category, only real skills listed
[ ] 3-4 featured projects with concrete descriptions
[ ] Contact info is easy to find
Pinned Repos (up to 6):
[ ] Each repo has a clear one-line description
[ ] Each repo has 5-10 relevant topics/tags
[ ] Each repo has a README with install instructions
[ ] No forks, no tutorials, no abandoned projects
[ ] Pin order reflects your best work first
[ ] Pins collectively tell a coherent career story
Activity & Hygiene:
[ ] Private contributions enabled in settings
[ ] No committed secrets or API keys (audit with trufflehog or similar)
[ ] Recent activity visible (within last 30 days)
[ ] Commit messages are clear, not "fix" or "update"
"Audit my GitHub profile at github.com/[username]. Rewrite my profile README using the headline/about/stack/projects/contact structure. Recommend which 6 repos to pin and in what order, what topics to add to each, and what repo descriptions to update. Write the full README.md I can copy-paste, and flag any anti-patterns you see. Target audience: hiring managers looking for AI-native developers."
The agent becomes your profile strategist — turning a scattered GitHub presence into a portfolio that gets interviews.
| Skill | When to use it |
|---|---|
portfolio-strategy | Decide which projects to showcase and what story they tell together |
career-advisor | Full career lifecycle — self-discovery, gap analysis, job search, growth |
Create, update, close, and report on Jira tickets. Reads JIRA_URL and JIRA_TOKEN from environment variables or .env file. Provides comprehensive guidance for Jira ticket lifecycle management, API integration, and automation workflows.
AI-powered clarity engine for job seekers and founders. Not a resume fixer — a thinking partner. Helps you find your north star sentence: who you are, what you do, and why it matters to someone else. Produces one core version and targeted variations for different audiences. Use before interviews, investor conversations, networking events, or any moment where you need to introduce yourself under pressure.
Review code for bugs, security issues, and improvements. Use when the user asks to: (1) review code or a diff, (2) check code quality, (3) find bugs or issues, (4) audit for security problems. Trigger on phrases like 'review this', 'check my code', 'any issues with this', 'code review'.
Full-lifecycle career advisor — self-discovery through career growth. Use when the user wants to: (1) discover identity, values, ikigai, or skills inventory, (2) define dream job or analyze career gaps, (3) close gaps via learning, certifications, portfolio, or personal branding, (4) build narrative hats or role-specific positioning, (5) search for jobs, generate resumes, or prepare for interviews, (6) onboard into a new role, track achievements, or generate self-reviews, (7) backfill past roles, run weekly recall, or check career readiness. Proactively nudges users to document achievements NOW — not when they need a resume. Consolidates the former career-resume, career-tracker, portfolio-strategy, and github-profile skills.
Synthesizes professional accomplishments into high-converting landing page content. Triggers when a user wants to build a landing page, launch a consulting service, productize their career expertise, create a service page, or monetize their skills. Also triggers on: "freelance", "consulting offer", "offer page", "service offering", "turn my experience into a business". Uses career-advisor data as the foundation.
Query Google NotebookLM notebooks directly from Claude Code for source-grounded, citation-backed answers. Use when user mentions NotebookLM, shares a notebooklm.google.com URL, asks to query their notebooks/docs, or says "ask my NotebookLM", "check my docs", "query my notebook". Provides browser automation, notebook library management, and persistent auth. Drastically reduced hallucinations through document-only responses from Gemini.