pp-recipe-goat

Pass

Audited by Gen Agent Trust Hub on May 15, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONDATA_EXFILTRATION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill installs a CLI binary and an MCP server from the vendor's own repositories using npx and go install. Specifically, it references the @mvanhorn/printing-press package on NPM and the github.com/mvanhorn/printing-press-library repository on GitHub.
  • [COMMAND_EXECUTION]: The skill facilitates the execution of the recipe-goat-pp-cli binary using the Bash tool, passing user-supplied arguments and subcommands to the binary.
  • [DATA_EXFILTRATION]: The CLI includes a --deliver webhook:<url> feature that allows the output of any command to be POSTed to an arbitrary external URL. While a functional part of the tool, this represents a potential path for data transmission to external endpoints.
  • [PROMPT_INJECTION]: The skill processes data from external sources (USDA API and 37 third-party recipe websites), which introduces a surface for indirect prompt injection.
  • Ingestion points: Results from recipe searches and USDA nutrition lookups (SKILL.md).
  • Boundary markers: No explicit delimiters or boundary instructions are used when interpolating external data.
  • Capability inventory: Access to Bash and Read tools; the CLI can write to files and send webhooks.
  • Sanitization: No specific sanitization or filtering of the ingested external content is described.
Audit Metadata
Risk Level
SAFE
Analyzed
May 15, 2026, 05:49 PM