Time-Lapse Album Custom Summarization Skill | 时光相册自定义浓缩分析技能

Generates condensed album highlights based on specified keywords or targets. It extracts specific target segments from long videos and compiles them into an...

This skill looks like a legitimate client for a remote 'timelapse summarization' API, but there are important mismatches you should understand before installing or running it: - SKILL.md forbids reading local memory and mandates a strict open-id retrieval process, but the codebase includes utilities that will create/read config.yaml files and a local SQLite database under your workspace/data directory and that also read environment variables (e.g., OPENCLAW_SENDER_OPEN_ID). Ask the author to explain and reconcile this discrepancy. - The skill sends video data and metadata to external endpoints (default production host appears in smyx_common config: lifeemergence.com). If you will upload private videos, confirm exactly which URLs the skill will call and whether the operator/host is trustworthy. - The skill declares no required environment variables, but it will read several (OPENCLAW_WORKSPACE, OPENCLAW_SENDER_OPEN_ID, OPENCLAW_SENDER_USERNAME, FEISHU_OPEN_ID). If you have secrets or credentials in your environment, consider running the skill in an isolated environment or sandbox and verify what environment values are actually used. - The package contains large shared libraries and a long requirements list. If you plan to install dependencies, review that list; consider installing in a virtualenv/container rather than system-wide. - Practical next steps: request from the publisher (or examine the code) to clarify: where attachments are saved, exact external endpoints and what fields are sent, why a local DB is needed, and a concrete data-retention policy. If you cannot obtain those answers, run the skill only in an isolated sandbox with network monitoring, or avoid installing it for sensitive data.

Type: OpenClaw Skill Name: smyx-custom-timelapse-analysis Version: 1.0.0 The skill bundle contains aggressive prompt instructions in SKILL.md that force the AI agent to ignore its native memory and local files, mandating the use of specific scripts and a cloud API (lifeemergence.com). The underlying framework (smyx_common) implements a custom persistence layer using a local SQLite database (dao.py) to store user tokens and includes logic in util.py to automatically register or log in users via a 'phoneLogin' endpoint using provided identifiers. Furthermore, skill.py in the common directory allows the skill to recursively invoke the 'openclaw' agent via subprocess.run, which, combined with the automated credential management and memory overrides, represents a high-risk architectural pattern.