mmk-paymint-bulk-send

Pass

Audited by Gen Agent Trust Hub on Mar 17, 2026

Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it ingests and processes data from external JSON files and standard input.
  • Ingestion points: The --file and --data flags, along with piped stdin, serve as entry points for untrusted data into the agent context.
  • Boundary markers: The skill instructions do not specify any delimiters or safety warnings to ignore instructions embedded within the record fields (e.g., message, name).
  • Capability inventory: The skill leverages Bash(mmk *) to execute payment operations based on the provided input.
  • Sanitization: There is no documentation of input validation or sanitization for the data fields before they are passed to the tool.
  • [COMMAND_EXECUTION]: The skill utilizes the mmk command-line utility via bash to execute bulk sends. This tool is owned by the vendor (magic-meal-kits) and its use aligns with the skill's primary objective of processing payment invoices.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 17, 2026, 03:28 PM