{"id":77094,"plugin_id":"plugins_6ab42163e9048191ab6e8c14258a48c3","kind":"skill","collection_source":"plugin_package","comparison_source":null,"observed_at":"2026-11-03T00:12:26.244Z","digest":"6c4cdcda158a1537285b434f7abebd096e67daee4c674226ac4d2dadb0633781","against":null,"payload":{"description":"Compare RAG retrievers, vector stores, rerankers, frameworks and models on grounded technical criteria rather than hype.","included_files":[],"name":"compare-rag-stacks","skill_md_contents":"---\nname: compare-rag-stacks\ndescription: Compare RAG retrievers, vector stores, rerankers, frameworks and models on grounded technical criteria rather than hype.\n---\n\n\n# Compare RAG technology choices\n**Trigger:** Choosing managed vs self-hosted retrieval, embeddings, vector DB, query rewriting, rerankers, frameworks or providers.\n\n1. Confirm user workload and constraints: corpus shape, security model, scale, freshness, latency, price ceiling, stack/operator skill, portability and failure tolerance.\n2. List candidate options with clearly sourced capability facts. Verify current pricing, releases, quotas and API changes on official sites rather than guessing.\n3. Compare on security/ACL enforcement, data residency, ingestion behavior, filtering, hybrid retrieval, operational complexity and measured latency/quality where evidence exists.\n4. Design a fair benchmark with fixed corpus/eval set and budget, not an unsupported numeric ranking.\n5. Recommend one default with trade-offs, conditions that change the choice, a reversible pilot and acceptance gates.\n\nRead `../../references/technology-comparison.md`, `../../references/architecture-patterns.md`.\n"},"changes":[],"summary":"First saved snapshot. No earlier version is available for comparison.","summary_kind":"deterministic","summary_metadata":{}}