Compress CLI output and cache contextual data for AI-assisted coding sessions.
npx skills add https://github.com/yvgude/lean-ctx --skill lean-ctx
Please help me install this Agent Skill. Skill: lean-ctx Source: https://github.com/yvgude/lean-ctx/tree/main/skills/lean-ctx Command: npx skills add https://github.com/yvgude/lean-ctx --skill lean-ctx
LeanCTX tackles the issue of bloated LLM prompts and repetitive context by compressing CLI output and providing a persistent cross-session memory layer for developers.
Install LeanCTX and run lean-ctx -c <command> to compress verbose output and preserve context for AI-assisted coding.
High-intent search queries and answers about installing and using this skill.
You can compress LLM context by running a command like `lean-ctx -c <command>` to intercept and compress verbose CLI output before it reaches the AI. This reduces token costs and preserves context for AI-assisted coding.
Cross-session memory works by caching contextual data and applying LITM-aware positioning to maintain relevant context across different chats and projects. This prevents repetitive context loading and optimizes attention for iterative coding workflows.
Yes, context compression integrates directly with Cursor and Claude Code. The system provides simple agent hooks to connect and cache contextual data across these AI coding platforms, accelerating development without altering existing workflows.
Yes, the system supports tree-sitter parsing for 14 programming languages to extract meaningful APIs and structures. This enables accurate dependency graphs and delta updates during context compression across various coding projects.
Yes, a Rust binary is required to run this context compression system. The binary powers the core compression engine, shell hook integration, and the context server's 24 intelligent tools for cached reads and task-aware relevance scoring.
yellow-seedPort Dockerfile RUN and ENV steps into portable shell scripts for Codex environments.
TheModdersDenDiagnose AI token usage and apply tiered optimization strategies to conversation history.
yankeeinlondonExtract symbols, imports, and exports from source files using Tree-sitter.
omeragaakbasDiagnose context stuffing in AI workflows and identify structural fixes.
