AI Engineering

Navigator Intent Brief renders a one-screen scope brief for ambiguous task prompts before code changes. It confirms goal, scope, limits, and verification, and pauses for up to two open questions or drift checks.

Navigator Knowledge Graph queries and manages a project knowledge graph across tasks, SOPs, memories, and concepts. Use it to find related knowledge, capture patterns or pitfalls, and initialize or inspect the graph.

Navigator Profile manages saved user preferences and learned corrections for bilateral modeling. Use it when a user asks to remember preferences, view their profile, or after repeated corrections.

Optimizing Bundles Forge improves bundle-plugin or single-skill projects by tightening triggers, lowering token use, fixing audit findings, and refining workflows. Use it when iterating on a skill or bundle-plugin from feedback or analysis.

Qdrant Advisor loads the latest official Qdrant skills to diagnose deployment, search, scaling, memory, and client SDK issues. Use it when a Qdrant cluster, collection, or vector-search question needs current guidance.

Qdrant Clients SDK provides official client libraries for Python, JavaScript, Rust, Go, .NET, and Java. Use it to connect applications to Qdrant through REST or gRPC and to fetch code snippets for common tasks.

Qdrant Edge guides building with Qdrant’s embedded in-process shard. Use it for local sync patterns, BM25 and hybrid search on device, snapshots, and operating Edge correctly in Python or Rust.

Qdrant Latency Remote helps diagnose search latency spikes in Qdrant, such as p99 regressions over time. Use it to inspect what changed and narrow down likely causes with skills.qdrant.tech.

Qdrant Model Migration guides changing embedding models in Qdrant without downtime. Use it to re-embed data, add or switch vector fields, or run old and new models side by side.

Qdrant Multitenancy guides how to isolate tenant data in Qdrant for multi-user search and RAG systems. It covers payload partitioning, custom sharding, collection-per-tenant cases, and data residency planning.

Qdrant Search Quality diagnoses bad search relevance, low recall, and low precision in Qdrant. Use it to identify whether the model, the data, or retrieval settings are causing poor matches and to measure retrieval quality.

Qdrant Search Quality Diagnosis helps isolate why retrieval results are irrelevant, missing, or low-recall. Use it to compare exact vs approximate search, check quantization and filtering, and measure retrieval quality with labeled baselines.