Scraper
PassAudited by ClawScan on May 1, 2026.
Overview
This is a straightforward scraping helper for user-authorized pages that fetches URLs and saves cleaned text locally, with its main risks clearly disclosed.
This skill appears safe for its stated purpose. Before installing, be aware that it can fetch supplied URLs and save cleaned page contents locally, so use it only for public or authorized pages and avoid storing sensitive content unless you intend to keep it in the local OpenClaw workspace.
Findings (2)
Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.
The agent can make web requests to URLs it is given, which may contact sites or services on the user's behalf.
The script fetches a user-supplied URL directly. This is expected for a scraper, but users should ensure the URL is public or authorized because the code does not itself enforce that boundary.
parser.add_argument("--url", required=True, help="Public URL to fetch") ... urllib.request.urlopen(req, timeout=20)Use only with public or explicitly authorized URLs, and do not use it for pages behind logins, paywalls, captchas, robots restrictions, or rate limits.
Scraped text and source URLs may persist locally and be visible to later workflows that inspect this workspace.
The skill stores extracted page content and job metadata persistently in a local OpenClaw memory/workspace directory. This is disclosed and purpose-aligned, but saved content can remain available after the task.
All outputs are stored locally under: `~/.openclaw/workspace/memory/scraper/jobs.json` ... `~/.openclaw/workspace/memory/scraper/output/`
Avoid saving sensitive page contents unless you are comfortable keeping them locally, and delete the scraper memory/output directory when the saved data is no longer needed.