volcengine-tos-vectors-skills

**Judgement:** This skill is a solid, production-oriented integration for Volcengine TOS Vectors, well-suited for teams already on Volcengine who need a managed vector database for semantic search, RAG, and recommendation workloads. **Strengths:** - Covers the full lifecycle: bucket/index management, vector CRUD, and KNN similarity search with metadata filters. - Good performance practices (batch limits, dimension guidance, choice of distance metrics) and clear service limits. - Documentation and examples align directly with common AI patterns (semantic search, RAG, recommendations). **Key Risks / Limitations:** - Tightly coupled to Volcengine TOS Vectors; not suitable if you need a cloud-agnostic or self-hosted vector store. - Requires correct configuration of credentials, endpoint, and region; misconfiguration or missing env vars will cause failures. - Operational behavior (retries, latency, availability) is largely dependent on the underlying TOS Vectors service and SDK, not the skill itself. **Recommended Scenarios:** - You are already using Volcengine and want a managed vector DB for embeddings, semantic search, or RAG. - You need to manage vector indexes and perform metadata-filtered KNN search at moderate to high scale with clear operational limits. - You want a straightforward way to plug TOS Vectors into LLM-based workflows (e.g., context retrieval, personalization, or recommendations).