# MCP Projects at AI Tinkerers

> Canonical HTML: https://aitinkerers.org/technologies/mcp
> Markdown URL: https://aitinkerers.org/technologies/mcp.md
> Technology record last updated: 2026-09-18T15:14:00Z
> Generated: 2026-09-20T23:37:22Z

MCP is the open-source standard for securely connecting AI agents (like LLMs) to external tools, data, and enterprise workflows.

The Model Context Protocol (MCP) functions as a standardized integration layer: think of it as a USB-C port for AI applications. Developed and open-sourced by Anthropic, this protocol allows large language models (LLMs) to access real-time context and execute actions via external tools like GitHub, Jira, or proprietary databases . It uses a simple JSON-RPC interface to define tools, schemas, and endpoints, which enables AI agents to perform complex, state-changing tasks—such as creating a GitHub issue or running a test script—rather than just generating text . MCP is essential for building agentic AI systems that can autonomously pursue goals and operate within defined safety and permission boundaries .

- Official technology site: https://modelcontextprotocol.io/
- Public AI Tinkerers demos and talks: 129
- Result page: 1 of 6

## Recent Public Talks and Demos

### [AI Engineer Kids Day — Hands-on AI, Coding and Emerging Tech for Ages 9–15](https://paris.aitinkerers.org/talks/rsvp_vzUScVFw9t0)

AI Engineer Kids Day is a hands-on learning experience for kids aged 9–15, built around two parallel technical workshops led by Cassandra Chin and Henry Collie, Technical Curriculum Developer at Neo4j. Cassandra Chin will lead a hands-on AI gaming workshop where participants use Mistral and Godot to build and customize a simple video game, then connect it to an MCP server that tracks scores. Henry Collie will lead Phippy’s AI Friend, a workshop where participants help Phippy the giraffe on a rescue mission while wiring sensors on a breadboard and writing simple code with an Arduino Nano. Along the way, they explore foundational concepts behind how modern applications run and communicate. Both workshops are designed for beginners and make AI, coding, connected systems, and modern software concepts accessible through practical building. The workshops will be accessible in both English and French. We are currently recruiting volunteers to help provide live interpretation between the two languages so that children and parents can participate comfortably regardless of their preferred language.

- Event context: AI Engineer Kids Day — Paris (Ages 9–15) — 2026-09-20 — Paris
- Public talk page: https://paris.aitinkerers.org/talks/rsvp_vzUScVFw9t0

### [Cut your AI and SaaS bill in half](https://barcelona.aitinkerers.org/talks/rsvp_andGu4eTve0)

We built Tarvis, an interface for deploying and managing self-hosted applications, AI tools, and coding workflows across your own servers. I’ll show the working system live, starting with power of self-hosting and self-hosted tools like SearXNG connected to coding agents through MCP and a containerized agent harness with Git access. Then I’ll show how we use Tarvis to deploy and manage these applications, connect tools to different model providers, and show a few other tools that can help reduce the cost of both SaaS and AI.

- Event context: AI Tinkerers Barcelona - September Demo Night — 2026-09-17 — Barcelona
- Public talk page: https://barcelona.aitinkerers.org/talks/rsvp_andGu4eTve0

### [A Self-Improving AI OS for the Chronically Self-Employed](https://barcelona.aitinkerers.org/talks/rsvp_AA_yQ6WeHtc)

A personal AI operating system, built as a plain-text git repo, that connects to my entire freelance business in real time: live contracts, open invoices, job feed, inbox, calendar, and meeting notes. It uses that context to find work, draft proposals, track clients, and run a morning brief every day without me asking. This system is built upon the foundation of Nate Herk's AIS-OS combined with Andrej Karpathy's LLM-wiki guidelines. For the demo I'll show it live: open a terminal, pull the job feed from Contra, trigger a skill, watch the wiki update, then walk through the self-audit loop that finds gaps in the system and ships one improvement per week. The repo stays visible throughout. No slides. We may jump into live client / prospects on Contra.

- Event context: AI Tinkerers Barcelona - September Demo Night — 2026-09-17 — Barcelona
- Public talk page: https://barcelona.aitinkerers.org/talks/rsvp_AA_yQ6WeHtc

### [SkillCheck - building linter for Agent Skills](https://copenhagen.aitinkerers.org/talks/rsvp_E9DOLVyNJX4)

MCP server checking Agent Skills for compliance with standards, semantic coherence, structure, quality patterns and more.

- Event context: September Demo Night — 2026-09-16 — Copenhagen
- Public talk page: https://copenhagen.aitinkerers.org/talks/rsvp_E9DOLVyNJX4

### [SamePage: you and your agent, literally on the same page](https://columbus.aitinkerers.org/talks/rsvp_7Rnho8suGUw)

SamePage is a Rust library for adding agents to a UI responsibly: they join over the AG-UI protocol with full tracing and human approvals, so you don't rebuild that plumbing for every app you want AI in. It runs at several tiers, from a loose brainstorm room up to a surface where sign-offs are enforced. The point is aligning intent. We work with an agent across sessions until the intent lands in a PRD, then implementation drifts from it, small at first, compounding. Load the PRD into SamePage and the agent displays its understanding as objects on the page. We sign off on each one, and the parts that seemed clear but were misaligned surface immediately. This is one of the very few arenas where human and agent are actually looking at the same thing. Each sign-off then compiles into a machine-checked requirement, so alignment is enforced for the life of the repo, not just the meeting. Live demo: a room started from the terminal with zero build step, a coding agent attached over MCP authoring UI while we talk, my marks on its work read back as meaning instead of coordinates, and the event log showing exactly who did what.

- Event context: AI Tinkerers - Columbus September Meetup — 2026-09-07 — Columbus
- Public talk page: https://columbus.aitinkerers.org/talks/rsvp_7Rnho8suGUw

### [Goals MCP for preventing agents from reporting "Fake Done"](https://missoula.aitinkerers.org/talks/rsvp_bd91Kk9nVv8)

Show case of the goals mcp ( https://github.com/brucepro/goals_mcp ) in real time on a project using claude code. The primary purpose of the MCP is to prevent an agent from providing a done message when the work has not been completed.

- Event context: AI Tinkerers – Missoula Inaugural Meetup · 26 August 2026 — 2026-08-26 — Missoula
- Public talk page: https://missoula.aitinkerers.org/talks/rsvp_bd91Kk9nVv8

### [How I built a Multi-Agent system from scratch](https://da-nang.aitinkerers.org/talks/rsvp_J6dVjfFty_w)

A multi-agent AI system with three context control mechanisms I discovered while building it: context enhancement, context shrinking, and unique tool result handling. The demo shows agents collaborating while these mechanisms silently manage context across handoffs.

- Event context: AI Tinkerers Da Nang: Inaugural Meetup — 2026-08-15 — Da Nang
- Public talk page: https://da-nang.aitinkerers.org/talks/rsvp_J6dVjfFty_w

### [Build and Share Apps as Easily as a Google Doc](https://atlanta.aitinkerers.org/talks/rsvp_V8A3ha9mDRc)

Cloud for small personal apps. Demo: building a real-time collaboration app with live audience participation.

- Event context: AI Tinkerers Atlanta x AI Collective: Community Demos at ATL Tech Week — 2026-08-13 — Atlanta
- Public talk page: https://atlanta.aitinkerers.org/talks/rsvp_V8A3ha9mDRc

### [Claude Code Forgets Between Sessions. I Built a Fix.](https://atlanta.aitinkerers.org/talks/rsvp_F2czpdWF3xs)

I built MLA, a context coordination layer that keeps Claude Code's working context consistent and up to date across sessions. The problem is simple: during one coding session, Claude discovers an important constraint, changes an architectural decision, or learns that an old assumption is wrong. Then you start a new session and that working context is gone. Teams try to preserve it in CLAUDE.md, documentation, memory files, rules, or other knowledge systems, but those systems eventually drift because staying current still depends on a human or agent remembering to make the update. I initially thought this was just my workflow. After talking with 30+ developers using coding agents, the same failure mode kept showing up. MLA maintains an active source of truth as coding work happens. It captures important decisions and discoveries, preserves where they came from, detects conflicting or outdated information, and keeps track of what is currently valid. When a new Claude Code or Codex session starts working, MLA gives it the relevant current context before it acts. I'll demo the full loop live: something changes during one coding session, the source of truth is updated, then a fresh session starts with an outdated assumption and MLA supplies the current information automatically. I'll also show the hooks, retrieval path, traces, and governance behind the handoff.

- Event context: AI Tinkerers Atlanta x AI Collective: Community Demos at ATL Tech Week — 2026-08-13 — Atlanta
- Public talk page: https://atlanta.aitinkerers.org/talks/rsvp_F2czpdWF3xs

### [Agents write the code, phones approve the merge: device-in-the-loop CI for AI-built mobile apps](https://seattle.aitinkerers.org/talks/rsvp_RZQnTBW084M)

An AI-assisted delivery workflow where coding agents ship mobile apps and physical phones give the final approval. A GitHub issue gets claimed by a Claude Code agent running in a long-lived tmux loop. The agent opens a PR with tests. CI on a self-hosted runner then installs the build on a real Pixel and a Wear OS watch through custom MCP device-control servers, runs the acceptance check, and posts device screenshots back to the PR. Live: the loop end to end on real hardware, MCP tool-call traces, the CI wiring, and the watchdog code that keeps a multi-day agent loop alive. Prerecorded backup for the device segment in case venue wifi dies.

- Event context: AI Dev Tools Track - Seattle - August 4 — 2026-08-05 — Seattle
- Public talk page: https://seattle.aitinkerers.org/talks/rsvp_RZQnTBW084M

### [What if loops are the problem?](https://nyc.aitinkerers.org/talks/rsvp_UEhibnXmrYg)

Loopbreaker is a public, local-first MCP server and visual SQLite review graph extracted from a deeper review-and-shipping system embedded in my application, Rordi. I built the original system after an AI code review reached thirteen passes and continued discovering new reasons not to ship. Loopbreaker isolates the reusable mechanism: an issue has a frozen set of enforced behaviors, evidence is attached to those behaviors, and review is limited to one comprehensive pass, one repair-verification pass, and—only when necessary—one shipping-decision pass. I’ll demo the working system live: clone the public repo, start its local MCP server, connect an AI coding agent, load a synthetic version of the thirteen-pass incident, and inspect the review graph. I’ll show the agent querying the acceptance surface, recording exact test evidence, completing a repair pass, and stopping automatically—while the visual interface separately explains whether the issue is actually ready to ship.

- Event context: NYC Summer Social: Rooftop Oyster Demo Day (two years of Tinkering in NYC) — 2026-07-29 — New York City
- Public talk page: https://nyc.aitinkerers.org/talks/rsvp_UEhibnXmrYg

### [AI Boost - a context sharing system](https://valencia.aitinkerers.org/talks/rsvp_Zd-yr9DNsxM)

AI Boost is a context sharing MCP that allows repeatedly used context to be easily searched for and added to the LLM

- Event context: AI Tinkerers Valencia July Demo Night — 2026-07-28 — Valencia
- Public talk page: https://valencia.aitinkerers.org/talks/rsvp_Zd-yr9DNsxM

### [Dig.rest](https://orange-county.aitinkerers.org/talks/rsvp_a0OCfbGWoFA)

Digging into events of the world, building cause and effect maps and getting to know the layers behind the news.

- Event context: AI Tinkerers Orange County: Tuesday, July 21, 2026 at Centercode — 2026-07-22 — Orange County
- Public talk page: https://orange-county.aitinkerers.org/talks/rsvp_a0OCfbGWoFA

### [The Completion Utility Stack: Launching Thousands of AI-Native Businesses for the Agentic Economy](https://orange-county.aitinkerers.org/talks/rsvp_g0ZRnVDKJI8)

I built NetShow IQ1, the full agentic operating stack from NetShow.AI for creating AI-first, AI-native businesses where digital crews move users from intent to completed outcome. IQ1 is designed around the Completion Utility: the idea that, just as electricity, water, and gas became foundational utilities for modern life, reliable task completion becomes a new utility for the agentic economy. In the live demonstration, I’ll show how IQ1 turns a request like “I need this handled” into an orchestrated workflow across virtual agents, tools, memory, MCPs, skills, approvals, and reporting. I’ll walk through the working system, architecture, agent harness, workflow routing, tool calls, logs, and how a business-specific digital crew can be composed for real consumer and business services. NetShow.AI is building economic infrastructure for thousands of AI-first businesses and services across major categories of life, work, commerce, local services, operations, support, home, and environment.

- Event context: AI Tinkerers Orange County: Tuesday, July 21, 2026 at Centercode — 2026-07-22 — Orange County
- Public talk page: https://orange-county.aitinkerers.org/talks/rsvp_g0ZRnVDKJI8

### [Quality in the age of AI](https://islamabad-rawalpindi.aitinkerers.org/talks/rsvp_L3a9baolTt0)

Orcastrator is an agentic QA automation system built to help teams keep software quality aligned with the speed of AI-assisted development. I began by discussing the quality challenges the software industry is facing as AI accelerates the delivery of code and features. I highlighted the concerns being reported by technology companies and research teams, and why maintaining quality is becoming increasingly important as development continues to move faster. From there, I used a live demo application to show how quality automation can now be authored through just a few prompts. The demonstration focused on how teams can move from understanding what needs to be tested to creating and executing meaningful automated tests with far less manual effort. I then showed how we have approached this problem with Orcastrator. The platform enables agentic quality automation that can operate at a speed much closer to modern software delivery, while remaining structured and powerful enough to provide meaningful coverage. The demo showed how application areas, test cases, execution runs, logs, screenshots, traces, and product health reporting can be connected within a single quality model. This makes the process of creating, running, and understanding tests as seamless as possible without sacrificing the depth needed to identify genuine product risks. The broader point was that, as software delivery becomes increasingly agentic, quality assurance must evolve alongside it. Testing cannot remain a slow, disconnected stage at the end of development. It needs to become an integrated and equally agentic part of how software is built and delivered.

- Event context: AI Tinkerers Islamabad - July 18, 2026 — 2026-07-18 — Islamabad Rawalpindi
- Public talk page: https://islamabad-rawalpindi.aitinkerers.org/talks/rsvp_L3a9baolTt0

### [using AI to create lighthearted presentations for a powerpoint karaoke style event](https://nurnberg.aitinkerers.org/talks/rsvp_Z06wBRUAe1k)

setup to create satirical slides for events that help people with overcoming their anxiousness actually presenting their great stuff to a broader audience!

- Event context: AI Tinkerers Nürnberg: June Meetup (Community Hub) — 2026-06-24 — Nürnberg
- Public talk page: https://nurnberg.aitinkerers.org/talks/rsvp_Z06wBRUAe1k

### [Three Models, One Ledger: A Live Multi-Model Witness Protocol over MCP](https://austin.aitinkerers.org/talks/rsvp_NEzGU02MWI8)

A three-model witness protocol: to verify a piece of AI-assisted work, the same proof is sent to Claude, GPT, and Gemini independently over MCP, and it only seals into an append-only ledger when they agree. Live, I'll fire a witness request from my phone, show all three models attest in real time, and watch the record seal into the vault — the payloads, the disagreement handling, and the actual database row. No slides.

- Event context: AI Tinkerers Austin x PostHog Demo Night — 2026-06-18 — Austin
- Public talk page: https://austin.aitinkerers.org/talks/rsvp_NEzGU02MWI8

### [Finding problems with your agents in production](https://montreal.aitinkerers.org/talks/rsvp_B5uOVF3D5dE)

My apps have AI features, which I'm tracking in my observability tool, but they have bugs - how do I find them and fix them? I'll show you what our workflow is, at PostHog, to observe our AI features/agents, evaluate their output, surface problems, and fix them. We'll see how we find issues that we might have never found otherwise, in both our products and agents.

- Event context: AI Tinkerers Montreal - June Demo Meetup — 2026-06-17 — Montreal
- Public talk page: https://montreal.aitinkerers.org/talks/rsvp_B5uOVF3D5dE

### [Wrapping MCPs to beat context pollution](https://prague.aitinkerers.org/talks/rsvp_bShvtr6rBCs)

Working production system that uses clever caching to wrap MCP servers into subagents, reducing token counts by 95+% in tool descriptions and another 95+% in task output tokens that stay in the context window

- Event context: AI Tinkerers Prague June Meetup — 2026-06-12 — Prague
- Public talk page: https://prague.aitinkerers.org/talks/rsvp_bShvtr6rBCs

### [Inhabited-design: an adversarial-persona Claude Code skill that produces delightful AI design (not slop)](https://seattle.aitinkerers.org/talks/rsvp_r_8j0_bmQOg)

Inhabited-design is a Claude Code skill that fights the AI Slop problem: the purple gradients, 3 column CTAs, rounded buttons. It turns "build me an X for Y" into a delightful design with unique UI that actually has a personality. Instead of one-shotting that, the skill samples a different real designer to inhabit on every run. So an energy drink for finance bros came back as Peter Zumthor with actual references, then critiques its own work in the voice of another inhabited critic: Tanner, the 26 year old investment banker living in midtown. The skill uses two established techniques, verbalized sampling (https://arxiv.org/abs/2510.01171) and Intent Factored Generation (https://arxiv.org/abs/2506.09659) to coax the model out the same old design attractors that are models often gets trapped in when designing content. Burns a silly number of tokens. Turns out unique design is expensive. Code: github.com/Shimin-Zhang/inhabited-design

- Event context: AI Dev Tools Track - Seattle - June 8 — 2026-06-09 — Seattle
- Public talk page: https://seattle.aitinkerers.org/talks/rsvp_r_8j0_bmQOg

### [What Is Your Coding Agent Actually Searching?](https://columbus.aitinkerers.org/talks/rsvp_t4IMPq7lJnc)

I built SearchBench, a harness for running controlled evaluation rounds over coding-agent search behavior. A SearchBench round takes real bug-localization tasks, gives an agent access to a repository, and checks whether it found the files that were actually changed in the human fix. The harness compares an incumbent search strategy against one or more challengers, then writes a static evidence bundle with exact-hit, hop-distance, token-usage, failure, and report artifacts. For the demo, I’ll run a small live ablation round over three cases. I’ll reuse cached Bash/native-search results as the incumbent, then run a few IC challenger policies in parallel. The question is: when IC succeeds or fails, was the important factor anchor quality or graph lookahead? I’ll show the actual workflow: preflight cost prediction, run planning, parallel execution, actual spend, cost-prediction error, generated bundle artifacts, and a short report explaining what changed. One thing I want to show explicitly is how much information the harness can derive from a small, bounded run: exact hits, hop distance, token usage, failure modes, and the next optimization target.

- Event context: AI Tinkerers - Columbus June Meetup — 2026-06-01 — Columbus
- Public talk page: https://columbus.aitinkerers.org/talks/rsvp_t4IMPq7lJnc

### [How We Built AI Agents That Buy Media Across 10+ Ad Platforms](https://seattle.aitinkerers.org/talks/rsvp_6LZbpmkAStQ)

Synter is an agentic AI platform that autonomously manages paid media campaigns across Google, Meta, LinkedIn, TikTok, Reddit, X, and Microsoft Ads. I'll demo the live system: how agents create campaigns, adjust bids, allocate budgets, and generate ad creatives in real time — no human clicking required. We'll look at the agent execution loop, the MCP server that lets Claude and Cursor control ad accounts directly, and the cross-channel dashboard that unifies data from all platforms.

- Event context: GTM / Growth Track - Seattle — 2026-05-27 — Seattle
- Public talk page: https://seattle.aitinkerers.org/talks/rsvp_6LZbpmkAStQ

### [Building an AI publishing platform with an AI dev squad](https://tokyo.aitinkerers.org/talks/rsvp_CzSJETAJ9-4)

CloudAuthor — a full-stack AI authoring platform, 14 agents, 27 skills, 300+ tests, built in 2 months on .NET Aspire and TypeScript/Vite. Live demo in two halves. First, I'll run the end-to-end publishing pipeline through CloudAuthor's agents: ideation → research → quote-finding → drafting → editing → banner creation → Japanese translation → carousel generation. Each stage is a specialized agent handing off to the next, all streaming in front of you. Then I'll switch to Visual Studio Code and show the squad of 7 specialized coding agents I actually use to build CloudAuthor itself — the agents that wrote the agents.

- Event context: AI Tinkerers Tokyo - Shinagawa: May 26th Meetup — 2026-05-26 — Tokyo
- Public talk page: https://tokyo.aitinkerers.org/talks/rsvp_CzSJETAJ9-4

### [Conversation as a Signal Query Engine - HQIQ Maestro](https://dubai.aitinkerers.org/talks/rsvp_2-90lqKKdGc)

Maestro / Hadi is a voice - first generative UI agent platform — no chat bubble, no text input. A full-screen avatar composes the interface in real time as the user speaks: scenes transition, listings render, comparison views compose themselves, contextual sheets open. The agent IS the frontend. We have shipped Translator Mode — a conversational capability where the same agent enters bilingual interpreter persona mid-session when asked. A realtor speaks English, a guest speaks Arabic (or any of 30+ languages Gemini Live supports), the avatar mediates the exchange in real-time bidirectional audio, then exits gracefully when the realtor asks to return to normal — with all prior property context preserved. Architecture: two function tools (enable_translator_mode / disable_translator_mode) inject interpreter instructions into the agent's chat context via update_chat_ctx, and emit TRANSLATOR_MODE data packets that surface a UI status indicator on the frontend. The LLM detects the trigger phrase naturally — no regex, no state machine. Provider-agnostic: runs on Google DeepMind's Gemini Live API for native audio, or OpenAI Realtime as a fallback. Litmus test: "Would this have been impossible with a chat interface?" Yes. Voice translation IS the conversational cadence chat removes. The translator pattern generalizes — same hook supports accessibility narrators, interview coaches, any role-shaped persona switch. The agent's persona is its capability, not its costume.

- Event context: AI Tinkerers Dubai - May Demo Day — 2026-05-23 — Dubai
- Public talk page: https://dubai.aitinkerers.org/talks/rsvp_2-90lqKKdGc

## Related Technologies

- [Claude Code](https://aitinkerers.org/technologies/claude-code) ([Markdown](https://aitinkerers.org/technologies/claude-code.md)) — 214 public demos
- [TypeScript](https://aitinkerers.org/technologies/typescript) ([Markdown](https://aitinkerers.org/technologies/typescript.md)) — 205 public demos
- [Python](https://aitinkerers.org/technologies/python) ([Markdown](https://aitinkerers.org/technologies/python.md)) — 662 public demos
- [Claude](https://aitinkerers.org/technologies/claude) ([Markdown](https://aitinkerers.org/technologies/claude.md)) — 173 public demos
- [OpenAI API](https://aitinkerers.org/technologies/openai-api) ([Markdown](https://aitinkerers.org/technologies/openai-api.md)) — 520 public demos
- [Gemini](https://aitinkerers.org/technologies/gemini) ([Markdown](https://aitinkerers.org/technologies/gemini.md)) — 188 public demos
- [Docker](https://aitinkerers.org/technologies/docker) ([Markdown](https://aitinkerers.org/technologies/docker.md)) — 147 public demos
- [LangChain](https://aitinkerers.org/technologies/langchain) ([Markdown](https://aitinkerers.org/technologies/langchain.md)) — 445 public demos
- [Node](https://aitinkerers.org/technologies/node) ([Markdown](https://aitinkerers.org/technologies/node.md)) — 99 public demos
- [Playwright](https://aitinkerers.org/technologies/playwright) ([Markdown](https://aitinkerers.org/technologies/playwright.md)) — 32 public demos
- [GitHub](https://aitinkerers.org/technologies/github) ([Markdown](https://aitinkerers.org/technologies/github.md)) — 74 public demos
- [LLM](https://aitinkerers.org/technologies/llm) ([Markdown](https://aitinkerers.org/technologies/llm.md)) — 123 public demos
- [Cursor](https://aitinkerers.org/technologies/cursor) ([Markdown](https://aitinkerers.org/technologies/cursor.md)) — 65 public demos
- [Next](https://aitinkerers.org/technologies/next) ([Markdown](https://aitinkerers.org/technologies/next.md)) — 186 public demos
- [GitHub Actions](https://aitinkerers.org/technologies/github-actions) ([Markdown](https://aitinkerers.org/technologies/github-actions.md)) — 20 public demos
- [HTTP](https://aitinkerers.org/technologies/http) ([Markdown](https://aitinkerers.org/technologies/http.md)) — 14 public demos
- [OpenAI](https://aitinkerers.org/technologies/openai) ([Markdown](https://aitinkerers.org/technologies/openai.md)) — 112 public demos
- [PostgreSQL](https://aitinkerers.org/technologies/postgresql) ([Markdown](https://aitinkerers.org/technologies/postgresql.md)) — 105 public demos

## More Results

- Next: https://aitinkerers.org/technologies/mcp.md?page=2