v0.5.0 - Cross-platform generic MCP install, per-agent AI onboarding guide

What's new

• Cross-platform, machine-agnostic install: the launcher probes the installed Microsoft Edge or Google Chrome on Windows, macOS and Linux, with a PATH fallback; no Windows-only executable path is hard-coded.
• Configurable browser engine, CDP port, dedicated profile directory and auto-launch switch, overridable through UPLUS_BROWSER, UPLUS_CDP_PORT and UPLUS_PROFILE_DIR; the legacy auto_launch_edge config field maps automatically onto the new generic settings.
• rename.md: a step-by-step playbook any MCP-compatible AI agent follows after installation (environment check, install, client registration, two-mode sign-in, patrol, objective-question answering, human takeover).
• Login detection prefers the session token before page text, fixing a first-navigation hydration timing miss that could report a signed-in session as logged out.
• README, README.zh and docs/mcp-clients.md are generalized for Windows / macOS / Linux first-time users, with bare Python and uplus-mcp console-script registration options.

Tooling & quality

• 28 tools: 10 uplus_* business tools and 18 browse_* generic browser tools.
• 73 unit tests pass; a clean virtualenv install was verified; a real-platform read-only regression (courses, homework, experiments, attachment question classification) passed without answering or submitting anything.

Scope & integrity

• Objective questions (single choice, multiple choice, true/false, fill-in) can be filled and submitted after the model's answers are reviewed; subjective answers, programming, attachments, experiment reports and group deliverables are always handed back to the user.
• The server stores no passwords, cookies or session snapshots in the repository.

Getting started

Follow the Quick start in README.md, then register the stdio server in your MCP client; read rename.md for the AI onboarding script.