Knowledge Share
Technical articles, tutorials, and insights
LLM-Driven Automated Testing Series (04): Web UI Automation — Midscene's Vision-Driven Scripting
Midscene, open-sourced by ByteDance, takes the pure-visual-locating path: no selectors at all — just hand a screenshot to a vision model and describe 'click the submit button' in natural language. This post breaks down its three model-role split (Default/Planning/Insight), aiAct's replanning loop, the distinct invalidation logic behind planning-cache and locate-cache, and where the robustness/cost trade-off actually lands compared to DOM/accessibility-tree-based approaches like Stagehand.
LLM-Driven Automated Testing Series (03): Unit Test Generation — TestGen-LLM and Qodo Cover
Tell an LLM to 'just write some tests' and you usually get code that doesn't compile, asserts something trivially true, or contributes nothing to coverage. Meta's TestGen-LLM paper doesn't try to make the model smarter — it bolts on a mechanical filter pipeline that discards anything that fails to prove its own worth. This post breaks down that filter design and how the open-source implementation, Qodo Cover, turned it into a runnable CLI tool.
LLM-Driven Automated Testing Series (02): The Unified Technical Foundation — Visual Grounding / DOM Semantics / Computer Use
The same 'find the submit button and click it' task gets solved three completely different ways across open-source projects: dedicated visual grounding models, DOM/accessibility-tree structured understanding, and raw coordinate-clicking Computer Use. This post builds the shared vocabulary the rest of the series will reuse, and explains what benchmarks like ScreenSpot, ScreenSpot-Pro, and OSWorld actually measure.
LLM-Driven Automated Testing Series (01): Why Testing Is LLM's New Battlefield
Traditional automated testing has three long-standing problems: fragile locators, the oracle problem, and maintenance cost — and they reinforce each other. This opening piece breaks down the nature of these three problems, and explains why testing is one of the rare LLM Agent use cases with a 'clear right-or-wrong signal.' The next 13 posts will each dissect a representative open-source project's specific solution.
One Open Source Project a Day (No. 223): TeamAI-CLI — Tencent's Team-Level AI Agent Middleware, Turning Individual AI Skills Into Shared Team Assets
TeamAI-CLI is Tencent's open-source team collaboration middleware. It isn't another AI coding assistant — it's a meta-management system that lets already-installed tools like Claude Code, Codex, and Cursor share skills, rules, and knowledge bases across a team. It syncs team standards through a push/pull workflow, builds a codebase knowledge graph, and uses friction-driven signals to capture and share hard-won experience automatically. 4.8k Stars, MIT License.
One Open Source Project a Day (No. 224): ARTEMIS — Google's Open-Source Mobile AI Automation Framework, Letting AI Assistants Operate Phones Like a Human
ARTEMIS is Google's open-source Android automation framework that turns natural-language instructions into reliable mobile automation. It offers two execution modes — Flash (fast reactive) and Pro (multi-agent planning and verification) — and integrates natively with Claude Code, Antigravity, Codex, and other AI IDEs via MCP, achieving 99%+ task completion on the AndroidWorld benchmark. 8.1k Stars, Apache-2.0.
One Open Source Project a Day (No. 225): NanoJev — A 0.6B Parallel Decision Model That Outputs Probability Distributions Instead of Generating Tokens
NanoJev is a lightweight recreation of the "Jev" parallel decision model system. Built on a Qwen3-0.6B backbone with dedicated decision heads, it takes states and questions as input and outputs complete probability distributions directly — with zero output-token decoding. It supports three decision head types (dynamic choice, boolean, ordered score) and has been validated on maze navigation and Snake game benchmarks. 954 Stars, MIT License.
One Open Source Project a Day (No. 226): AIO Sandbox — Browser, Shell, File, MCP, and VSCode Server in One Docker Container
AIO Sandbox (agent-infra/sandbox) is an all-in-one AI agent sandbox environment that unifies browser automation, a shell terminal, file operations, MCP servers, and VSCode Server in a single Docker container. Its core design is a unified filesystem shared across all tools. It ships official Python/TypeScript/Go SDKs, four built-in ready-to-use MCP servers, and has verified integrations with Browser Use, LangChain, and OpenAI Assistants. 6.0k Stars, Apache-2.0 License.
Open Source Project #222: security-audit — Cloudflare's Skill That Turns Your Coding Agent into a Six-Phase Security Auditor, 13k Stars
Cloudflare's open-source coding-agent skill that turns your agent into a security auditor. Six-phase workflow: reconnaissance, coverage-led hunting, candidate validation, structured output, independent record verification, and target-neutral reporting. Key design: adversarial validation (the checker is never the finder), machine-readable findings.json with zero-dependency validators, additive multi-run coverage. JavaScript, MIT, 13k Stars.
One Open Source Project a Day (No. 221): Coder — Self-Hosted Cloud Dev Environments, Now Also a Safe Workshop for AI Agents
Coder is a self-hosted cloud development environment platform that defines workspaces with Terraform across AWS/Kubernetes/Docker, cutting onboarding time from days to seconds. Its newer AI Agent capability runs the agent's reasoning loop on the control plane instead of inside the workspace, paired with an AI Gateway for centralized model governance, cost tracking, and auditing. 14.9k Stars, AGPL-3.0.
One Open Source Project a Day (No. 220): WeKnora — Tencent's Enterprise Knowledge Framework, From RAG Q&A to a Self-Evolving Wiki
WeKnora is Tencent's open-source LLM knowledge framework, combining fast RAG Q&A, multi-step ReAct agent reasoning, and a Wiki mode that auto-generates an interlinked knowledge base. Supports 20+ LLM providers, 10+ document formats, and enterprise-grade multi-workspace RBAC — it's the core technology behind the WeChat Dialog Open Platform. 25.4k Stars, MIT License.
One Open Source Project a Day (No. 217): free-for-dev — The 137k-Star Ultimate Vault of Free Services for Developers
free-for-dev is one of GitHub's highest-starred resource repositories, cataloging 60+ categories and hundreds of SaaS/PaaS/IaaS services with genuine free tiers — from major cloud platforms to CI/CD, databases, AI tools, and auth services. 137k Stars, maintained by 1,600+ contributors, the must-bookmark "freebie guide" every developer needs.