Analyze code repositories and generate reports with file:line pointers and diagrams.
npx skills add https://github.com/uv-xiao/pkbllm --skill uv-repo-analysis
Please help me install this Agent Skill. Skill: uv-repo-analysis Source: https://github.com/uv-xiao/pkbllm/tree/main/human/repo-analysis Command: npx skills add https://github.com/uv-xiao/pkbllm --skill uv-repo-analysis
This Skill provides a comprehensive, actionable analysis of a code repository, enabling rapid understanding of its architecture, components, and LLM-specific implementation details.
Run the init_repo_analysis.py script with the URL of the repository you want to analyze.
High-intent search queries and answers about installing and using this skill.
To analyze a code repository, you can generate reports that map architecture, identify workflows, and isolate key components. The tool produces actionable diagrams and file:line pointers, helping you quickly understand overall structure and data flow.
To onboard to a new LLM serving framework, you need a deep dive into its prefill/decode loops, KV cache, scheduler, and sampling. A targeted repository analysis isolates these specific implementation details, allowing you to become productive much faster.
Yes, you can generate architecture diagrams and module tables for a codebase. The analysis provides a clear overview of system structure, detailing installation processes, entry points, and configuration workflows.
The best way to identify extension points and data flow is to run an automated repository analysis script. It deeply analyzes the code to highlight key functions, runtime loops, and specific areas for feature extension, providing actionable reports.
No specific dependencies are required to run the repository analysis script. You simply execute the init_repo_analysis.py script with the URL of the repository you want to analyze to start generating reports.
NOMARJDetect malicious patterns in repositories, packages, MCP servers, and AI skills.
karrtik159Generate optimized multi-stage Dockerfiles with layer caching and security practices.
simoabidMeasure Core Web Vitals and analyze runtime and bundle sizes with Lighthouse CLI.
practical-stackAnalyze LLM agent and plugin repositories to extract architecture, patterns, and insights.
