Orchestrate semantic search across code, docs, tasks, and memory using Reciprocal Rank Fusion.
npx skills add https://github.com/lev-os/lev-content --skill lev-find
Please help me install this Agent Skill. Skill: lev-find Source: https://github.com/lev-os/lev-content/tree/main/sources/skills/lev-find Command: npx skills add https://github.com/lev-os/lev-content --skill lev-find
This Skill provides a unified interface to search across diverse knowledge sources, including code, documentation, tasks, and memory, ensuring you can quickly find the information you need.
lev get "authentication patterns" to find relevant code, design documents, and previous discussions about authentication strategies.Use the lev get skill to search for information about error handling in the codebase.
High-intent search queries and answers about installing and using this skill.
To search across codebase and documentation simultaneously, this Skill orchestrates semantic retrieval across multiple indexes and fuses results using Reciprocal Rank Fusion. This unified interface ensures you quickly find relevant information across diverse knowledge sources.
The best way to find prior art and recall past discussions is using cross-domain contextual retrieval over sessions, tasks, and memory. It addresses the need for unified context gathering, ensuring you discover previous solutions within your development ecosystem.
Yes, you can use external research backends like Exa and Valyu for codebase search. The Skill supports these external backends to enable comprehensive discovery and iterative refinement across your knowledge sources.
Reciprocal Rank Fusion improves memory and task retrieval by orchestrating semantic search across multiple indexes and combining their rankings. This unified approach ensures you quickly find the exact information needed across diverse knowledge sources.
Yes, unified search works with ideas, sessions, and skills indexes. It seamlessly searches code, docs, sessions, tasks, and memory simultaneously, allowing you to gather relevant context for new features like authentication patterns.
dexhunterOrganize agent knowledge into PARA folders with semantic search.
zeikarRun parallel Codex and Claude research paths to produce structured pre-implementation reports.
orthogonal-shOrchestrate web and dataset searches across multiple backends with parallel API calls.
DrLuggelsFuse vector and keyword search results using Reciprocal Rank Fusion.