qwencloud-vision

[QwenCloud] Understand images and videos with Qwen vision models. TRIGGER when: user wants to analyze, describe, or extract information from images or videos...

This skill appears to be a normal QwenCloud vision client (OCR, visual reasoning, video analysis) and its bundled Python scripts use only the standard library, but there are a few points to consider before installing or running it: - Registry metadata mismatch: the skill manifest in the registry does not list the required API key(s), yet the SKILL.md and scripts expect DASHSCOPE_API_KEY / QWEN_API_KEY. Treat this as a red flag — confirm with the skill author or source why the registry omitted required credentials. - Credentials: the scripts will look for your DASHSCOPE_API_KEY (or QWEN_API_KEY) in the environment or a .env file. Do not paste your real API key into chat. If you create a .env file via guidance in the docs, put a placeholder first and replace it locally; prefer setting the env var in a secure process. Verify the code never prints raw keys (the skill states it masks keys, but you should review the scripts yourself if you plan to run them). - File and config changes: the skill provides an agent-compatibility helper that can search for and append entries to project/agent config files (e.g., CLAUDE.md, AGENTS.md). It claims to ask before modifying files, but ensure you approve any changes and back up your configs first. - Local file uploads: the scripts can upload local files to provider-managed temp storage (oss://) instead of base64-embedding them. Only enable uploading if you are comfortable sending those files to QwenCloud/DashScope. For sensitive images/videos, use local base64 or avoid uploading. - Run in a controlled environment first: if you are unsure, run the scripts in a sandbox or review the bundled Python files (they are included) before executing. Check any network endpoints used (defaults are DashScope/QwenCloud endpoints like dashscope-intl.aliyuncs.com). If you require tighter assurance, request the skill author to update registry metadata to declare required env vars and document any config file modifications. If you want, I can extract the exact places in the code that require DASHSCOPE_API_KEY/QWEN_API_KEY and list which files would modify or scan user config paths so you can review them before running.

Type: OpenClaw Skill Name: qwencloud-vision Version: 0.2.0 The qwencloud-vision skill bundle provides a robust interface for interacting with Qwen vision models for image/video analysis and OCR. The implementation in `qwencloud_lib.py` and `vision_lib.py` uses standard Python libraries to handle API requests, file uploads to temporary storage, and environment variable management. The bundle includes a specialized update-check mechanism (`gossamer.py`) that emits signals to the agent's stderr to suggest installing updates or sibling skills, but the instructions in `SKILL.md` explicitly require user consent before taking action. Security measures are included to prevent the accidental leakage of API keys in plaintext.