hybrid-knowledge-retrieval-builder

Build governed hybrid retrieval projections with lexical and semantic search stages.

Updated Jun 1, 2026
One-click install
npx skills add https://github.com/aurora-atoms/lattice --skill hybrid-knowledge-retrieval-builder
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: hybrid-knowledge-retrieval-builder
Source: https://github.com/aurora-atoms/lattice/tree/main/skills/hybrid-knowledge-retrieval-builder
Command: npx skills add https://github.com/aurora-atoms/lattice --skill hybrid-knowledge-retrieval-builder

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) and schemas (resource) components.

What problem does it solve?

This Skill addresses the challenge of building reliable, evidence-grounded retrieval systems that maintain strict governance, ACL compliance, and citation accuracy, preventing common RAG failure modes like hallucination or unauthorized data exposure.

Core Features & Use Cases

  • Governed Projection Building: Create rebuildable retrieval projections that integrate lexical and semantic search with deterministic fusion.
  • Citation & Provenance: Ensure every retrieved chunk is traceable to its original source, version, and access-controlled locator.
  • Use Case: Use this to design a retrieval pipeline for internal policy documents where you must guarantee that only authorized users see specific content and that every answer is backed by a verifiable source citation.

Quick Start

Use the hybrid-knowledge-retrieval-builder skill to design and validate a citation-ready hybrid retrieval projection for the provided knowledge corpus.

Frequently Asked Questions about hybrid-knowledge-retrieval-builder

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I build a hybrid retrieval pipeline with strict ACL filtering?

To build a hybrid retrieval pipeline with ACL filtering, you construct governed projections integrating lexical and semantic search stages with deterministic fusion. This enforces access control compliance and ensures only authorized content surfaces during enterprise knowledge queries.

What is governed hybrid retrieval and when do I need it for RAG?

Governed hybrid retrieval integrates lexical and semantic search with strict ACL filtering and citation resolution. You need it for enterprise RAG applications to prevent unauthorized data exposure and ensure every answer traces back to a verifiable source.

How do I ensure source provenance and citation accuracy in a retrieval system?

Ensure citation accuracy by building rebuildable retrieval projections that resolve citations to original sources, versions, and access-controlled locators. This traceable source provenance grounds retrieved chunks and prevents evidence-grounded query hallucinations.

Can I use this approach for enterprise policy document retrieval?

Yes, you can apply governed hybrid retrieval to design pipelines for internal policy documents. It guarantees authorized users see specific content while providing verifiable source citations for evidence-based question answering.

What is the best way to achieve deterministic fusion in hybrid search?

Achieve deterministic fusion in hybrid search by constructing rebuildable retrieval projections that blend lexical and semantic stages under strict governance. This approach satisfies requirements for independent evaluation of retrieval performance and deterministic outputs.

Why does my RAG system expose unauthorized data during searches?

RAG systems expose unauthorized data when retrieval lacks strict ACL filtering. Applying governed hybrid retrieval projections enforces access control compliance at the search stage, preventing unauthorized content from surfacing in generated answers.