seekdb

Operate seekdb via CLI commands and look up seekdb documentation. Use when: executing SQL, exploring table schemas, managing vector collections, registering...

This skill appears to do what it says (operate seekdb and fetch docs) but its runtime instructions ask the agent to discover connection info by reading environment files and local config/db files that are not declared in the manifest, and it suggests running a curl|bash installer hosted on an S3 URL. Before installing or using it: (1) don't run curl | sudo bash from unverified S3 URLs — prefer vetted packages or official release pages; (2) confirm whether the agent will have access to ~/.seekdb, .env, or other local files and only allow that in an isolated/test environment; (3) if you must connect to remote databases, avoid storing credentials in global env or allow-read locations and prefer explicit DSNs you control; (4) ask the skill author to declare required config paths/env vars and to remove instructions that implicitly read unspecified user files. If you need help assessing specific commands (e.g., the installer URL), provide them and I can inspect further.

Type: OpenClaw Skill Name: seekdb Version: 0.2.0 The seekdb skill is a comprehensive database management tool for AI agents, providing capabilities for SQL execution, vector search, and AI model registration. It includes robust safety features such as mandatory row limits, write-protection flags, and automatic masking of sensitive fields (PII) in query results. While it references high-risk installation patterns (curl|bash) and performs remote network requests to fetch documentation from GitHub (raw.githubusercontent.com), these actions are well-documented and strictly aligned with the tool's purpose as a database CLI and documentation assistant.