Visual Summarization Skill | 视觉摘要智述技能

What to check before installing or enabling this skill: - Review and trust the remote API endpoints in skills/smyx_common/scripts/config.yaml (base-url-open-api, base-url-health). The default config points to lifeemergence.com — confirm that is expected and trustworthy. - Inspect skills/smyx_common/scripts/util.py (RequestUtil) to confirm what HTTP calls are made, what headers/credentials are sent, and whether any unexpected data exfiltration could occur. - Be aware the code will read local config.yaml files and may create a SQLite DB under the workspace (workspace/data/*.db). If you need to avoid any local persistence, do not install or run this skill. - Note the registry metadata declares no required env vars, but the code reads OPENCLAW_SENDER_OPEN_ID, OPENCLAW_WORKSPACE and other env vars — supply or restrict these as appropriate and consider running in a sandbox to observe behavior. - The bundle contains a large 'common' dependency list and an unrelated face-analysis skill; if you only need minimal summarization, ask the author for a slimmer package or examine whether those components are necessary. - If you will allow autonomous invocation, prefer to run this skill with least privilege: restrict workspace path, run in isolated environment, and avoid giving API keys until you confirm endpoint behavior. If you want, I can: (1) summarize the external endpoints and files the code will read/write, (2) extract and show the RequestUtil/http call patterns, or (3) point out exact lines where env vars and file writes occur.