Back to search
Noah-W-AI
MCP serverGitHub
Noah W: AI portfolio. AI agents, MCP servers and automations I built with Claude and Claude Code.
Safety scan
A static text scan, not a guarantee. Select the badge to see what was scanned and every finding.
Checked today
Badge for your README
Markdown
[](/listings/inoahguy1--noah-w-ai)HTML
<a href="/listings/inoahguy1--noah-w-ai"><img src="/api/listings/inoahguy1--noah-w-ai/badge.svg" alt="findsafeskills safety rating" /></a>Install
i
What was scanned
The files the safety scan read, as written by the repository's author and shown as plain text. Lines with a finding are marked.
README.md · README · 33 lines
- # Noah W: AI portfolio
- I build AI agents and automations, mostly with Claude and Claude Code. I run
- [Truevine Automation](https://truevineautomation.com), my freelance business that builds AI
- tools for small businesses.
- I designed every project below and built it with Claude Code as my pair programmer. I set the
- goals and the rules, and I review and test every change. Each project has a write-up with what
- it does, how it works, and what it shows. The source code is private, and I'm happy to walk
- through it or give a live demo. All sample data is made up.
- ## AI projects
- | Project | What it does | Tech |
- |---|---|---|
- | [Lead Qualifier Agent](projects/lead-qualifier-agent.md) | A Claude agent that looks up a lead with tools, scores it against a rubric, returns checked JSON, and drafts a follow-up that a person approves. Includes evals. 116 tests. | Python, Claude API (tool use, structured output), Pydantic |
- | [Leads MCP Server](projects/leads-mcp-server.md) | An MCP server that gives Claude safe access to a lead CRM, built to handle agent retries, guessed ids and skipped steps. 42 tests. | Python, MCP SDK |
- | [Claude Code Team Workflow](projects/claude-code-team-workflow.md) | My setup for rolling out Claude Code to a team safely: permission rules, hooks that block destructive commands, protected secrets, an audit log, review agents and CI. 210 tests. | Claude Code, Python, GitHub Actions |
- | [Truevine Workspace](projects/truevine-workspace.md) | A self-hosted chat where people and AI agents share rooms. Mention an agent and it answers, running on a linked computer's own Claude Code. 72 tests. | Python (no dependencies), SQLite, JavaScript |
- | [Truevine Systems](projects/truevine-systems.md) | AI systems for small service businesses: voice note to finished report, intake and follow-up, and a company brain that answers from the business's own notes. 57 tests. | Python, SQLite, Obsidian, Whisper, Claude Code |
- | [Truevine CRM](projects/truevine-crm.md) | The CRM for my business: pipeline, calendar, website bookings, n8n hooks for AI follow-up, and safe contact enrichment. | Supabase, Postgres, TypeScript, JavaScript |
- | [Truevine website](projects/truevine-website.md) | My business website, built with Claude. Live at [truevineautomation.com](https://truevineautomation.com). | React, CSS, GitHub Pages |
- Test counts are from running each project's test suite. The tests run offline, with no API key.
- ## What I work with
- Claude API · Claude Code (hooks, skills, subagents, MCP) · MCP servers · AI agents and evals ·
- Python · JavaScript · TypeScript · Supabase / Postgres · n8n · GitHub Actions
- ## Contact
- [LinkedIn](https://linkedin.com/in/noah-williams11) · [truevineautomation.com](https://truevineautomation.com)
Ratings and comments
Sign in to rate this skill.
Sign inFree to use. If it’s helped, pass it on.