{"skill":{"slug":"kernelgen-flagos","displayName":"KernelGen FlagOS","summary":"Unified GPU kernel operator generation skill. Automatically detects the target repository type (FlagGems, vLLM, or general Python/Triton) and dispatches to t...","tags":{"latest":"1.0.0"},"stats":{"comments":0,"downloads":135,"installsAllTime":0,"installsCurrent":0,"stars":0,"versions":1},"createdAt":1774240400901,"updatedAt":1774240607448},"latestVersion":{"version":"1.0.0","createdAt":1774240400901,"changelog":"- Initial release of kernelgen-flagos as a unified GPU kernel operator generation skill.\n- Automatically detects repository type (FlagGems, vLLM, or general Python/Triton) and dispatches to the appropriate specialized workflow.\n- Combines four sub-skills: kernelgen-general, kernelgen-for-flaggems, kernelgen-for-vllm, and kernelgen-submit-feedback, removing the need for separate installations.\n- Enforces mandatory use of the MCP tool for all operator code generation.\n- Supports user-invoked feedback and bug report submission via the bundled feedback sub-skill.\n- Enhances user experience with clear routing, explicit detection rules, and a quick reference guide.","license":"MIT-0"},"metadata":null,"owner":{"handle":"wbavon","userId":"s179mcxns07mhashdtzg4tbf6584wt8v","displayName":"Flagos","image":"https://avatars.githubusercontent.com/u/263577448?v=4"},"moderation":null}