Knowledge Share
Technical articles, tutorials, and insights
LLM-Driven Automated Testing Series (14): Production Engineering — Why Testing Turns Out to Be the Cheapest Line Item in This Whole Series
A code-generation workflow I documented earlier burned thousands of dollars in tokens during debugging. The testing scenarios verified across this series — AutoRestTest testing a 15-operation API for about a dime, pytest-triage capping itself at 10 model calls per run by default — are off by orders of magnitude. This article puts the evidence from 13 prior articles side by side to answer the three questions the planning doc left for last: why testing-scenario cost is what it is, where the human-in-the-loop gate actually belongs, and which scenarios should explicitly avoid LLMs altogether.
LLM-Driven Automated Testing Series (13): Flaky Tests Aren't a Rerun-Count Problem, They're a Classification Problem
pytest-rerunfailures (477 stars) and flaky (397 stars) handle test instability with the same one-liner: if it fails, rerun it a few times. That doesn't answer what the planning doc actually asks — was this failure a real bug, environment jitter, or a badly designed test? The only project actually attempting that root-cause classification with an LLM is a 2-star, less-than-two-months-old repo called pytest-triage — and its most instructive feature turns out to be the four safety invariants it wraps around AI judgment, guaranteeing AI is never allowed to affect whether a test actually passes or fails.
One Open Source Project a Day (No. 259): Modern Web Guidance — Chrome's Modern Web Best Practices for AI Coding Agents
Modern Web Guidance is a set of Agent Skills maintained by the Chrome, Edge, Firefox, and MDN teams, built to fix AI's habit of writing front-end code the old way. An agent searches a curated set of guides offline through a CLI (about 250 modern web features and 203 real-world use cases) and pulls code patterns, gotchas, and fallbacks into its context. The project ships its own closed-loop evaluation, and its published guided-versus-unguided pass rates improve by about 27 to 36 percentage points. Code is Apache-2.0, guide content is CC-BY 4.0, about 4k Stars.
One Open Source Project a Day (No. 260): Plannotator — An Annotatable Review Desk for AI Coding Agents, for Plans and Code Changes Alike
Plannotator is a local, browser-based review surface that works with ten agents, including Claude Code, Codex, Copilot CLI, Gemini CLI, OpenCode, Kiro, Droid, Amp, DeepSeek Harness, and Pi. When an agent proposes a plan or finishes a change, a browser UI opens; you annotate it paragraph by paragraph, approve or send it back, and the feedback returns to the agent in one click. It covers plan review, Markdown and web annotation, local and PR code review, an inbox, and encrypted sharing. Dual-licensed Apache-2.0 or MIT, about 10k Stars.
One Open Source Project a Day (No. 261): Uptime Kuma — A 90k-Star Self-Hosted Monitoring Tool, One Docker Command to Replace Uptime Robot
Uptime Kuma is a self-hosted monitoring tool from louislam, with a polished UI and an easy setup. It supports ten monitor types including HTTP(s), TCP, keyword, JSON query, Ping, DNS, Push, and Docker containers, checks as often as every 20 seconds, ships with more than 90 notification services, and offers status pages that can be mapped to your own domain. One docker run command is enough to deploy. MIT License, about 94k Stars.
Is Loop Engineering a New Technology, or Just Reheated Leftovers? Explained Through Training a Dog Classifier
Everyone in AI is talking about loop engineering. Take it apart and it looks a lot like the most basic machine learning training loop - a goal, an evaluation, feedback, something adjustable, and a cycle that runs until the goal is met. This article starts from a dog-recognition model, then walks through three real agent tasks (a large code migration, customer support ticket handling, a competitor research report) to show how a loop is built, why it raises the success rate, and whether it is really new.
Open Source Project #258: Harness² — Google Research's Recursive Agent Harnessing: Keep the Weights Fixed, Optimize the Shell Around the Model
Google Research's open-source framework for automatically optimizing an agent harness: no model training, just a task-level recursion — Propose, Probe, Compose — over eight editable components (system prompt, guardrails, memory, tool guidance, subagents, skills, scripts, plugins), plus a persistent improvement harness that accumulates lessons. On JobBench, LAB, and WorkBuddy-Bench, a single step lifts GLM-5 and Gemini 3.5 Flash above their base harnesses. Apache-2.0, 23k+ Stars.
LLM-Driven Automated Testing Series (12): Beyond Pixel Diffs — Who's Actually Making LLMs 'Understand' UI Changes
BackstopJS, at 7.2k stars, handles rendering noise with exactly four numeric config knobs — misMatchThreshold, requireSameDimensions, ignoreAntialiasing, usePreciseMatching — which is still just fighting font-rendering jitter with pixel-level tolerance. The project that actually wires a VLM/LLM into the visual-regression judgment loop is a 23-star repo called vlmkit, whose own internal benchmark notes state outright that one model hallucinates red into red. Putting these two side by side answers, concretely, what semantic visual regression actually solves — and what new problems it introduces.
One Open Source Project a Day (No. 256): Anatomy — Have Claude Invent a Machine for Any Technical Concept and Draw It as an Interactive Isometric Illustration
Anatomy is an Agent Skill. Give it a technical concept and it invents a physical device whose behavior embodies that concept (a cache key becomes a pin-tumbler lock, a shared GPU context becomes a gantry plotter), then drives the picture from a small model to produce an interactive isometric SVG illustration, with optional WebGL shaders and 3D. A dependency-free drawing kit, a nine-step workflow, and built-in geometry and overlap audits. Works with Claude Code, Codex, OpenCode, and more. MIT License, about 25k Stars.
One Open Source Project a Day (No. 255): Mika — An AI Agent for Android That Operates Your Apps and the Web on a Hidden Background Screen
Mika is an open-source Android AI assistant. You give it one instruction, and it looks at the screen, finds buttons, taps, and types like a person across apps and the web, on a hidden background display so your main screen stays free. There is no Mika server; requests go only to the model provider you configure. It ships presets for DeepSeek, Qwen, Kimi, Claude, Gemini, OpenAI, Ollama and more, plus a built-in browser, scheduled tasks, a Linux sandbox, and MCP/CLI access. Written in Kotlin. Apache-2.0 License, about 7.2k Stars.
One Open Source Project a Day (No. 257): AI Design Skills — Design Guidelines Written as Rules an AI Can Execute, for Landing Pages and Figma
AI Design Skills is a set of design Skills from elayadesign. Each Skill is a self-contained folder, and there are two so far - landing-page-design (the full landing page flow from strategy and copy to visual system) and figma-designer (structured Figma work using variables, text styles, auto layout, and component states). The rules go down to font allowlists, spacing steps, easing curves, and banned words. Works with Claude Code, Cursor, Codex, Windsurf, and more. MIT License, about 14k Stars.
One Open Source Project a Day (No. 254): NVIDIA Auto Ontology — A Certified Semantic Layer for Data Agents That Says 'I Can't Answer That' When It Can't
Auto Ontology is NVIDIA's open-source Python framework that builds an ontology of terms, SQL attributes, and table relationships from enterprise metadata, stores it in PostgreSQL + pgvector, and serves it to AI agents through MCP, a CLI, and a REST API. Human-certified definitions drive answers, and when the retrieved candidates can't support a query it declines instead of guessing. Supports 10 databases. Apache-2.0 License, about 5.7k Stars.