paprika

Pass

Audited by Gen Agent Trust Hub on Jun 22, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill documentation instructs users to install the paprika-recipe-cli package from the NPM registry to provide the necessary functionality.
  • [COMMAND_EXECUTION]: The skill invokes the paprika command-line tool using various subcommands and arguments. Some arguments, such as recipe names in paprika recipe "<name>" and search terms in paprika search "<query>", are derived from user input.
  • [PROMPT_INJECTION]: The skill processes external data retrieved from the Paprika service, such as recipe instructions and grocery items. This constitutes an indirect prompt injection surface where the agent could potentially encounter and follow instructions embedded within the recipe content.
  • Ingestion points: Data returned by paprika recipes, paprika recipe, paprika meals, and paprika groceries (referenced in SKILL.md).
  • Boundary markers: None present; external content is processed directly.
  • Capability inventory: The skill is limited to executing the paprika CLI tool for data retrieval and display.
  • Sanitization: No explicit sanitization or filtering of the retrieved recipe or list content is mentioned.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 22, 2026, 03:16 PM
Security Audit — agent-trust-hub — paprika