I made three AI seats argue over the spec before any code. Here's what 10 runs showed.
Most of the bad code I've gotten from coding agents wasn't badly written. It was the right code for...
Tutorials
I Fed My AI Agent a Poisoned Log Line — It Tried to Hijack Itself in 4 Minutes
Last week I read about a lab experiment where a prompt injection was hidden inside an ordinary...
Baton: Who Keeps the Promise When the Shift Ends?
“Shift’s over. Who keeps the promise to call the customer back?” That question is where Baton...
Claude Code Diff Verification in 2026: How to Make Your Agent Assert Its Own Edits Did Not Break the Build
Stop trusting AI edits on faith. Learn how to build post-edit verification hooks that make Claude Code run git diff, tests, and type checks before you ever see the changes, turning...
I built LLM-Surgeon: A Zero-Dependency CLI combining CanIRun.ai + Artificial Analysis + llmfit for Local AI Hardware Diagnostics
Running local LLMs via Ollama, LM Studio, or llama.cpp usually starts with the same guessing game:...
Building Your First Model Context Protocol (MCP) Server: A Team Task Tracker with FastMCP and SQLite
Model Context Protocol (MCP) has completely changed how we think about AI integration. Instead of...
Putting AI Agents on Call: How We Built an Autonomous SRE War Room in Slack with Switch and Wispr Flow
Putting AI Agents on Call: How We Built an Autonomous SRE War Room in Slack with Switch and...
LifePilot: Building a Personal Life Manager Inside a Chat Room
Flip the Switch Hackathon 2026 A build log on three agents, one human veto, and what happened when...
The Judge Reads Tokens: An Adversarial Evaluation of an Eval Toolkit's Own Scorers
The Judge Reads Tokens: An Adversarial Evaluation of an Eval Toolkit's Own...
The Test Passed. The Coupon Was Still Wrong.
Building SYNKAI with Switch: a coding agent, an independent verification boundary, and a human...
How I audit AI-written code — and the bug that proved why
I ship as an AI-augmented solo builder, but I never trust what the model produces. Here's the review discipline that caught a missing Android permission in every earlier release bu...
The gate that stops an agent's memory from re-firing an attack
The carryover problem Most prompt-injection work treats the attack as a one-turn event....
Stop Rubber-Stamping Terraform: How to Build an LLM Plan Reviewer in CI
Every plan, read fully — including the Friday 5pm ones. Pull up a chair, grab some chai, and let...
Uneven Branches Make a LangGraph.js Join Node Run Twice. Super-Steps Explain Why
One branch has one step, the other has two, both feed combine. LangGraph.js runs combine twice. Why, and the two fixes: array edges and defer: true.
A human was about to approve the payment. Then your context compaction ran.
Automatic context compaction and human approval gates are on a collision course. A one-line entry in the OpenAI Agents SDK changelog names the bug, and the fix is an approval recor...
Why I Log the Reason, Not Just the Result
I run a scheduled agent that writes and publishes dev.to articles twice a day from a side project...
499 requests called themselves AI bots on our site in 14 days. We could verify 105.
Six AI bot names in 14 days of logs on one site: 105 of 499 requests verified, 176 from outside the operator's IP list. How to check your own.
If a Linter Can Check Your AGENTS.md Rule, the Rule Doesn't Belong in AGENTS.md
Claude Code's docs call CLAUDE.md context, not enforced config. Move every rule a linter, hook or deny rule can check. 40 rules become 9.
So you're interested in autonomous AI?
I read the systems, security and software engineering literature before designing the next version of...
Nobody had to ask for the logs
keryx has no users but me, and it still got seven bug reports in two days, all filed by one of my own agent sessions. Checking one meant reading exactly what the reporter had typed...
Your LLM Agent Is an Untrusted Employee: 6 Security Guardrails We Built for AI in Banking Operations
Prompt injection, hallucinated numbers, runaway costs and leaked card data: practical guardrails for putting AI agents into a security-sensitive platform.
A client's vibe-coded app was leaking data. Here's how I closed it in 24 hours
A client's internal app (an inventory dashboard connected to 4 live Shopify stores) had been built...
The coding-agent toolbox around the model: 8 notes on CLAUDE.md, mcp2cli, OpenCode and worktrees
Once you use a coding agent every day, the model itself stops being the main lever. What changes the...
AWS's Kiro Bots Start in Trust Mode, and Crew's Default Sandbox Leaves ~/.ssh Visible
Kiro bots went GA on November 5. New bots act without per-action approval, and Crew's default sandbox leaves ~/.ssh and ~/.aws open to agent shell commands.