Compresses code from 16 languages into token-efficient, fast-queryable summaries for AI agents.
npx skills add https://github.com/lev-os/lev-content --skill llm-tldr
Please help me install this Agent Skill. Skill: llm-tldr Source: https://github.com/lev-os/lev-content/tree/main/sources/skills/lev-find-llm-tldr Command: npx skills add https://github.com/lev-os/lev-content --skill llm-tldr
This Skill tackles the challenge of AI agents struggling with large codebases by providing highly compressed and rapidly searchable code intelligence, drastically reducing token usage and query times.
llm-tldr to get a concise, relevant code context, significantly speeding up the debugging process.Use llm-tldr to get a summary of the main function in the current project.
High-intent search queries and answers about installing and using this skill.
To reduce token usage during codebase analysis, you can compress code into structured formats using AST analysis and call graphs. This approach achieves up to 95% token savings, enabling efficient semantic search and debugging across 16 languages without hitting context limits.
The best way to speed up semantic search and debugging is generating compressed code intelligence via program slicing and data flow analysis. This method delivers queries 300x faster than raw code analysis, providing concise contextual insights for AI agents.
Yes, AST analysis and call graphs can optimize refactoring tasks across 16 programming languages. By extracting structured code intelligence and control flow data, AI agents query relevant contexts rapidly, bypassing the need to process raw codebases entirely.
Token optimization for code analysis works without hitting context limits by compressing codebases into fast-queryable intelligence structures. Analyzing control flow and data flow reduces the token payload by 95%, ensuring AI agents stay within operational context windows.
Querying program slicing and data flow logic for debugging requires pre-computed code intelligence structures. By compressing the codebase into structured formats, AI agents retrieve specific control flow paths instantly, bypassing thousands of lines of raw code.
tianzecnAutomate deep code analysis across multi-file codebases with semantic search.
johnlindquistSearch code with ripgrep and analyze results with Gemini to resolve software bugs.
Jurel89Guides developers through a structured verification process for code changes.
petrkohutCompress documents into dense, lossless distillates for LLM consumption.