Stop being your AI's eyes and hands: pwa-debug lets Claude debug your live, logged-in browser
If you've debugged a web app with an AI assistant, you know the loop: The AI asks what the console...
Tutorials
A 9B assistant built from our specs went from 6 to 32 right answers out of 32. Most of the fixes were code, not prompts
A personal assistant is easy to demo and hard to trust. The demo ends when the assistant answers....
The Two Workflows That Wrote to the Same Row and Both Said Success: Race Conditions in n8n
This is part of my build-in-public series on running LLM workflows in n8n in production. Everything...
Why I chose plain Markdown over a database for AI agent memory
Every time I opened a new Claude Code session, I spent the first few minutes explaining my work...
Playwright MCP snapshots cost me 22,556 tokens per page. Here's what I measured
If you let Claude Code, Codex or Cursor drive a browser, a big part of your token bill is page...
Build a Q&A Bot over Your Company Docs with Pinecone + Claude (Python, SDK v10)
Chunk, embed, store, retrieve, rerank and answer with citations: a complete RAG walkthrough on Pinecone's new document indexes and the Claude Messages API.
The A2A CLI and Rust: The Same Code, Three Ways to Serve It
A tour of the A2A CLI against Rust agents on A2A v1.0: the official Rust SDK, a plain stdin/stdout program wrapped by a2a server --exec, and an ADK coordinator on Gemini, each time...
Point Claude Code at axe-core's Violations, Not at 'Make This Page Accessible'
'Make it accessible' invites aria-label sprinkle and gives you no proof. Hand the agent axe-core's violations by rule id and a test that fails CI.
AI Execution Governance, Part 10: Can Revocation Stop a Call That Is Already Running?
A tool call has passed the gate and is running. Now the role is revoked. Does it stop, or does it...
How uv add silently downgraded my pypdf security patch (and how I fixed it)
I merged two PRs to fix my dependency alerts on GitHub. Both had empty diffs. GitHub kept showing 11...
My x402 API made $0.08 in its first week. Here is who actually came.
A few weeks ago I put a small API online that agents pay per call with x402, USDC on Base. It scans...
LLM Routing: Stop Paying Premium Prices for Simple Prompts
Someone asked: can we end the Globex contract early? Seven words. The question looked short and...
We Checked 61 Official MCP Servers Against the Auth Spec
Most SaaS companies now ship an official remote MCP server, and most of them sit behind OAuth. We...
My scheduled task didn't run, and the AI guessed the Mac was asleep without checking the logs
A daily scheduled task stopped twice. The first time, the AI guessed a cause without opening the logs. What I ask for now, and a small check for the run that never happened.
OpenAI Says ChatGPT’s Search-Like Query Volume Is About One-Fifth of Google’s
OpenAI says ChatGPT’s web-search-like query volume is roughly 20% of Google’s traditional search...
I ran the same multi-agent experiment twice. Half the conclusions flipped.
In August I ran a controlled experiment on my open-source multi-agent pipeline: the same task, done...
Autonomous Deep Research for AI Agents: Eliminating Hallucinated Citations and 85% Token Burn in Python
Building autonomous deep research pipelines (like Stanford STORM or OpenAI Deep Research) typically...
Building Model Context Protocol (MCP) Servers in 5 Lines of Python — The FastAPI of MCP
Anthropic's Model Context Protocol (MCP) has rapidly become the universal open standard for...
How to Run 30B MoE Models on an 8GB Consumer GPU with Hybrid Offloading in Python
Running massive 30B+ Mixture-of-Experts (MoE) models locally usually requires dual RTX 3090/4090s...
Indirect Prompt Injection: How to Protect Your AI Agent with Spotlighting and Prompt Shields
While my AI agent was reading a web page, the page tried to give it orders. The page started with...
Anthropic's FlashProve Proves Which Model Answered You, at 70-95x vLLM's GPU Time
Anthropic open-sourced a prototype that proves a reply came from committed weights. It runs on Llama 4 Maverick, not Claude, in a 1,024-token window.
How I Overhauled DocsGPT's Ingestion Engine & Supercharged RAG Context: My Hacktoberfest 2026 Journey
How I Overhauled DocsGPT's Ingestion Engine & Supercharged RAG Context: My Hacktoberfest 2026...
Switch from zero: ship a human-in-the-loop data quality agent to your team chat
I spent the Flip the Switch hackathon building Shield, an agent that checks a daily table, explains...
SynapseHealer: Building an Autonomous, Self-Healing Interoperability DLQ with InterSystems IRIS AI Hub
Built for the InterSystems Programming Contest: AI Agents in Interoperability Workflows (2026) ...