GitHub Repo Briefing
CloakBrowser
A source-patched Chromium build intended as a drop-in Playwright or Puppeteer browser for controlled browser testing and automation.
A first test
Run the supplied browser example against a local test page. Review the security, compliance, and site terms first; do not use it to bypass access controls or anti-abuse systems.
AiToEarn
A Chinese and English platform for using AI to create and distribute content across several publishing and marketing services, with plugin and MCP integrations.
A first test
Connect only a test account and publish one private draft. Check the generated copy, platform permissions, and deletion path before linking a production social account.
agentmemory
A persistent memory service for coding agents and MCP clients. It stores project context and exposes a demo server for checking how memory is added and retrieved.
A first test
Run the setup and agentmemory demo --serve in a disposable project. Store one fact, retrieve it in a new session, and verify the record before adding private notes.
financial-services
Claude plugins and reference material for financial analysis, investment research, due diligence, and other financial-services workflows.
A first test
Install one relevant plugin and test it with a public company filing. Check every number against the source document and keep the result as analysis support, not financial advice.
UI-TARS-desktop
A multimodal desktop agent stack with a CLI, web UI, browser and computer-use tools, MCP support, and model adapters for visual interaction.
A first test
Install the desktop client and use it on a test account. Ask it to open one local application and report the screen action sequence before allowing clicks that change data.
9router
A local AI router that presents one endpoint for several coding clients and providers, translates request formats, and can fail over between configured models.
A first test
Configure one provider and one low-risk coding request. Inspect the route, model, token count, and error fallback in the logs before adding more providers or secrets.
academic-research-skills
Imbad0202/academic-research-skills
A Claude Code skill workflow for academic research: plan the question, gather sources, write, review, revise, and finalize with a human in the loop.
A first test
Install it and run /ars-plan for one narrow question. Check the source list and claims after each stage; do not accept a final draft that lacks evidence for a key statement.
bun
A JavaScript and TypeScript runtime that also provides a package manager, bundler, test runner, and script launcher in one executable.
A first test
Install Bun in a sample project, run bun install, then run one script and one test. Compare the result with the project's current Node command before changing the build tool.
skills
A collection of small engineering skills for planning, implementation, testing, review, simplification, and release work. The files are editable and can be installed as a managed Claude Code plugin.
A first test
Install one skill set, not both installation methods. Run its planning and testing commands on a small change, then keep only the files that improve your actual review evidence.
CodeWhale
A terminal coding-agent harness that lets users bring their own model and run an open-source agent workflow from a local project.
A first test
Install it with npm install -g codewhale and run codewhale exec in a disposable clone. Give it a test failure and inspect the command history and patch before merging.
AI-Trader
An agent-native trading platform that exposes market data and trading workflows through skills and an API.
A first test
Use the paper or dry-run mode with historical data. Check the data timestamp, order assumptions, and risk limits; never connect a live brokerage account for a first test.
RuView
A WiFi-sensing system that estimates presence, movement, and some vital-sign signals from radio data without using a camera.
A first test
Run ruview doctor and ruview verify in a test room. Compare the output with a known movement test and treat any health-related signal as experimental, not a medical measurement.