rlm

SKILL.md

Recursive Language Models (RLM) CLI - enables LLMs to handle near-infinite context by recursively decomposing inputs and calling themselves over parts. Supports files, directories, URLs, and stdin.

Set an API key for your backend (openrouter is default):

| Directory | rlm ask . -q "..." | Recursive, respects .gitignore | | File | rlm ask main.py -q "..." | Single file | | URL | rlm ask https://x.com -q "..." | Auto-converts to markdown | | stdin | git diff \| rlm ask - -q "..." | - reads from pipe | | Literal | rlm ask "text" -q "..." --literal | Treat as raw text |

遞歸語言模型 (RLM) CLI - 使 LLM 能夠通過分解輸入並在各個部分上調用自身來遞歸處理大型上下文。用於代碼分析、差異審查、代碼庫探索。在“rlm 詢問”、“rlm 完成”、“rlm 搜索”、“rlm 索引”上觸發。 來源：rawwerks/rlm-cli。

查看原文