DuckDuckGo Search
Analysis
The skill appears purpose-aligned for web search and URL fetching, with no credential use or persistence, but users should notice its outbound web access and manual unpinned package install instruction.
Findings (3)
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.
Checks for instructions or behavior that redirect the agent, misuse tools, execute unexpected code, cascade across systems, exploit user trust, or continue outside the intended task.
parser.add_argument("url", help="URL to fetch") ... urllib.request.urlopen(req, timeout=timeout)The fetch helper makes outbound requests to a supplied URL. This is expected for a URL-fetching skill, but broad URL fetching can contact unexpected hosts if invoked with sensitive or internal URLs.
pip3 install duckduckgo-search
The documentation asks the user to install an unpinned third-party Python package. This is disclosed and related to the stated purpose, but users should verify the package source and version.
Checks for exposed credentials, poisoned memory or context, unclear communication boundaries, or sensitive data that could leave the user's control.
Supports text search with results including title URL and snippet. Also supports URL fetching to extract readable content from web pages.
The skill intentionally brings search snippets and fetched page text into the agent context. That content is untrusted internet data and could contain misleading instructions or prompt-injection text.
