Milvus Skill

Operate Milvus vector database with pymilvus Python SDK. Use when the user wants to connect to Milvus, create collections, insert vectors, perform similarity...

This skill is an instruction-only reference for using pymilvus and appears coherent with that purpose. Before installing or allowing autonomous use: (1) Understand that you will need to provide Milvus connection credentials (host/port, token or username/password) to actually run operations — only supply those to trusted agents or when you intend the agent to act on your database. (2) Examples show destructive DB ops (drop/truncate); confirm such actions with the agent before execution. (3) If you plan to use external embedding services (e.g., OpenAI) the skill's examples will require additional API keys and network access — those keys are not requested by the skill itself, so manage them according to your security policies. (4) Because the skill can perform admin-level DB operations when given credentials, limit its permissions (use least privilege) on production systems.

Type: OpenClaw Skill Name: milvus-skill Version: 0.0.2 The milvus-skill bundle is a well-documented set of instructions and references for managing Milvus vector databases via the pymilvus SDK. It includes proactive security guidance for the AI agent, such as explicitly instructing it to ask users for connection credentials rather than assuming or hardcoding them, and requiring confirmation before destructive operations like dropping collections. The provided Python patterns in SKILL.md and the references/ directory align with official library usage and do not contain any evidence of malicious intent, data exfiltration, or obfuscation.