Traction scoreGitHub stars can be faked, so popularity alone can be misleading. Traction Score looks for broader signs of recent attention, adoption, and active maintenance.
Waza provides a specialized health check for agent-assisted engineering environments. It helps AI automation builders identify when configurations for tools like Claude, Codex, or Pi have drifted or are being ignored by the model.
The skill focuses on the maintainability of the AI setup rather than application bugs, auditing MCP hooks and project constraints to ensure instructions remain reachable and effective.
Compatible AgentsThe repository documents support for these agents. The skills may also work with other agents that can load SKILL.md files, but they may need some setup or small changes.
Follow the documented setup, then try a first task.
I’m using
Shared Waza bundle via npx
Installs: Complete Waza bundle containing all eight documented skills; the shared directory may also be read by Gemini CLI, Copilot, Amp, Kimi Code CLI, and other compatible agents. · Claude Code, Codex, Cursor
Before you start
npx available in the terminal.
Run this command in a terminal. It installs the bundle globally for Claude Code, Codex, and Cursor, placing the shared copy in `~/.agents/skills`.
npx skills add tw93/Waza -a claude-code codex cursor -g -y
Installs: Complete Waza host plugin with host-namespaced skills; the supplied evidence does not establish exact Codex-plugin-to-SKILL.md path mappings. · Codex
Run this command in a terminal to add the Waza marketplace to Codex.
codex plugin marketplace add tw93/Waza
Then run this command in a terminal to install the full Waza plugin in Codex.
Installs: Waza ZIP skill bundle; the supplied evidence does not establish exact bundle-to-file mappings. · Claude Desktop
Download `waza.zip` from `https://github.com/tw93/Waza/releases/latest/download/waza.zip`. In Claude Desktop, open Customize > Skills > "+" > Create skill, then upload the ZIP.
Use the read skill to fetch and give me a concise, source-grounded summary of [URL or local PDF path]. Highlight the main points and state any paywall or extraction failure explicitly. Do not save a file.