Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
OpenClaw
Benign
high confidencePurpose & Capability
Name/description (网页剪藏) align with the included script and SKILL.md: the script extracts article content, saves to local Markdown, and has a Feishu save path. Optional Notion/pillow mentions exist but the script does not implement Notion or image-localization full logic — this is a minor mismatch but not suspicious.
Instruction Scope
SKILL.md instructs the agent to call scripts/clip_webpage.py to extract and save webpages. The script performs expected actions: HTTP GET of the provided URL(s), HTML parsing/readability extraction, optional API calls to Feishu if credentials and client library are present, and writes Markdown files to a user directory. Note: fetching arbitrary URLs means the agent's host will make outbound network requests to sites you supply (privacy/metadata consideration). The SKILL.md and script do not try to read unrelated system files or additional environment variables.
Install Mechanism
There is no formal install spec in the registry; SKILL.md recommends pip installing requests, beautifulsoup4, readability-lxml and optionally feishu-api, notion-client, pillow. This is a low-risk, common approach but relies on third-party Python packages — verify package names and sources before installing in a sensitive environment.
Credentials
Declared env vars (FEISHU_APP_ID, FEISHU_APP_SECRET, NOTION_TOKEN, CLIPPER_OUTPUT_DIR) are proportional and optional for the features described. The script only uses FEISHU_APP_ID/FEISHU_APP_SECRET when the Feishu path is selected; Notion support is listed but not implemented in the provided script (NOTION_TOKEN is unused). No unrelated credentials or high-privilege environment access are requested.
Persistence & Privilege
Skill is not always-enabled and is user-invocable; it does not request persistent system-wide privileges or modify other skills. It writes output files to a user directory (default ~/Documents/WebClips) which is expected for its purpose.
Assessment
This skill appears coherent for clipping webpages. Before installing: 1) Be aware the agent will perform HTTP GET requests to any URLs you provide — your host's IP and request metadata will be visible to those sites. 2) The skill writes files to ~/Documents/WebClips by default (CLIPPER_OUTPUT_DIR can be set). 3) Optional Feishu/Notion integrations require you to supply credentials — only provide these if you trust the feishu-api/notion-client packages and the environment; the script's Feishu function currently returns a simulated URL rather than a full implementation, and Notion support is not implemented in the included script. 4) The SKILL.md suggests running pip installs; review and install those packages in a controlled environment (or inspect them) before trusting this skill. If you need higher assurance, run the included script locally with test URLs and without providing any service credentials to verify behavior.Like a lobster shell, security has layers — review code before you run it.
chinesevk9775tge2a5zk7t8cg0wbkf0px851zepclippervk9775tge2a5zk7t8cg0wbkf0px851zeplatestvk9775tge2a5zk7t8cg0wbkf0px851zepproductivityvk9775tge2a5zk7t8cg0wbkf0px851zepscrapervk9775tge2a5zk7t8cg0wbkf0px851zepwebvk9775tge2a5zk7t8cg0wbkf0px851zep
License
MIT-0
Free to use, modify, and redistribute. No attribution required.