Web Browsing.Bak
PassAudited by ClawScan on May 1, 2026.
Overview
This appears to be a straightforward web search and page-fetching skill, with user-directed web requests and minor packaging metadata inconsistencies to notice.
This skill is reasonable for browsing and summarizing public web pages. Before installing, be aware that your searches and requested URLs go to external sites, and verify the package identity because the registry name/owner differs from the included metadata.
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.
Search queries and requested URLs may be sent to DuckDuckGo or the target website, and the target site will receive a normal web request.
The helper can fetch a caller-supplied URL and follow redirects. This is expected for a web-browsing skill, but it is a broad network action users should keep intentional.
response = requests.get(url, headers=headers, timeout=15, allow_redirects=True)
Only ask it to fetch URLs or search terms you are comfortable sending to external sites, and treat fetched webpage text as untrusted content.
The skill behavior looks coherent, but the mismatched labels may make it harder to confirm the exact publisher or package lineage.
The internal metadata names a different owner/slug than the registry listing for Web Browsing.Bak, which is a packaging/provenance inconsistency even though the included code is visible.
"ownerId": "kn7eqex1zb70x5x63754n9y2z182tm35", "slug": "web-browsing"
Verify that this is the intended package and publisher before installing, especially if you expected a different official web-browsing skill.