pp-roadside-america
Pass
Audited by Gen Agent Trust Hub on Sep 14, 2026
Risk Level: SAFEREMOTE_CODE_EXECUTIONDATA_EXFILTRATIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [REMOTE_CODE_EXECUTION]: The skill instructs the agent to install a CLI tool and an MCP server using
npx -y @mvanhorn/printing-press-libraryandgo install github.com/mvanhorn/printing-press-library/.... These resources are hosted on the vendor's official NPM and GitHub repositories and are necessary for the skill's operation. - [DATA_EXFILTRATION]: The skill documents the use of a
--deliver webhook:<url>flag within theroadside-america-pp-clitool. This feature allows the tool to POST command output to a user-specified URL. - [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data scraped from the web (RoadsideAmerica.com). Ingestion points: The CLI fetches HTML content from external web pages and parses it into structured formats (SKILL.md). Boundary markers: The skill does not explicitly mention delimiters or instructions to ignore embedded content in the scraped data. Capability inventory: The agent uses the Bash tool to interact with the CLI and local SQLite database. Sanitization: The CLI converts raw HTML into structured JSON or CSV, which provides a layer of data normalization.
Audit Metadata