ordercli

Pass

Audited by Gen Agent Trust Hub on Mar 30, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill installs the third-party ordercli tool from a non-standard GitHub repository (steipete/ordercli) using Homebrew or Go during the setup phase.
  • [COMMAND_EXECUTION]: Executes shell commands via the ordercli binary to manage logins, sessions, and food orders. This includes high-privilege operations such as importing Chrome browser cookies and accessing local application support folders to bypass bot protection.
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it ingests untrusted data from external food delivery APIs and restaurant menus.
  • Ingestion points: Fetches active order status, order history, and restaurant details from external Foodora and Deliveroo APIs (SKILL.md).
  • Boundary markers: No specific delimiters or boundary markers are defined to isolate external API data from the agent's instructions.
  • Capability inventory: The skill can execute external binaries, read/write configuration files, access browser session data, and perform financial transactions (reordering) (SKILL.md).
  • Sanitization: There is no evidence of sanitization or validation of the data retrieved from external sources before it is interpreted by the agent.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 30, 2026, 12:15 AM