stophy-web
Pass
Audited by Gen Agent Trust Hub on Oct 11, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill configuration in
SKILL.mdincludesnpx -y @stophy/cliin theallowed-toolssection, which downloads and executes the vendor's package from the NPM registry. - [COMMAND_EXECUTION]: The skill uses
Bashto execute shell commands such asstophy google searchandstophy google newsto perform search operations and manage local files. - [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from the web, including Google search results and academic papers, creating a potential surface for indirect prompt injection.
- Ingestion points: Data is retrieved from external web sources and saved to local JSON files (e.g.,
.stophy/search.json,.stophy/news.json) for processing by the agent as shown inSKILL.md. - Boundary markers: The skill does not define explicit delimiters or instructions to treat web-retrieved content as untrusted data.
- Capability inventory: The skill has access to shell execution (
Bash) and file writing capabilities. - Sanitization: No sanitization or filtering mechanisms for retrieved web content are mentioned in the skill's instructions.
Audit Metadata