Obsidian Clipper

Save web content (articles, videos, notes) to Obsidian vault with automatic classification, intelligent naming, and content extraction. Supports: 小红书 (Xiaoho...

Before installing or using this skill: - Inspect and edit the included config.json immediately. It ships with an absolute vault_path and collector_name belonging to the author; change the vault_path to your intended location or delete the file so the skill must prompt you on first run. - Understand that the skill writes Markdown files to whatever vault_path is present in config.json — test with a disposable folder first to confirm behavior. - Ensure the platform 'write' and 'web_fetch'/'web_search' tools it expects are the ones you trust; the SKILL.md also mentions browser snapshot and obsidian-cli (not declared in metadata). Install obsidian-cli only if you need CLI integration. - Because classification is silent by default, consider instructing the skill to prompt you for classification or review saved items until you’re comfortable with automatic behavior. - If you are concerned about autonomous invocation, consider disabling autonomous skill invocation or only enabling this skill when explicitly invoked. These steps reduce the risk of unexpected writes or privacy issues stemming from the packaged configuration and silent-save behavior.

Type: OpenClaw Skill Name: obsidian-clipper Version: 2.1.1 The obsidian-clipper skill is a functional tool designed to scrape web content and save it to a local Obsidian vault. It utilizes standard OpenClaw capabilities (web_fetch, web_search, and write) to extract information from platforms like Xiaohongshu, YouTube, and WeChat. The instructions in SKILL.md are focused entirely on content extraction, classification, and file management. While the included config.json contains a hardcoded absolute path specific to the author's local environment (/Users/lucien/...), this appears to be a packaging oversight rather than an intentional attempt at unauthorized file access or data exfiltration.