worker-image-build

Pass

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [COMMAND_EXECUTION] (LOW): The skill frequently executes shell commands via the gh (GitHub CLI) tool. It constructs these commands by interpolating user-provided inputs such as <CONFIG_NAME> and <RUN_ID>. This is necessary for the skill's primary function but requires the agent to be cautious of malformed input.
  • [EXTERNAL_DOWNLOADS] (SAFE): The skill interacts with the mozilla-platform-ops/worker-images repository. Although Mozilla is not on the provided list of trusted GitHub organizations, it is a highly reputable entity. The skill triggers remote workflows rather than downloading and executing binary artifacts locally.
  • [PROMPT_INJECTION] (LOW): The skill is vulnerable to indirect prompt injection (Category 8) because it processes untrusted user data without explicit sanitization or boundary markers.
  • Ingestion points: User input for <CONFIG_NAME> and <RUN_ID> in SKILL.md and references/examples.md.
  • Boundary markers: Absent; user input is directly concatenated into shell strings.
  • Capability inventory: Executes subprocess calls via gh workflow run, gh run watch, and gh run view.
  • Sanitization: Absent; the skill relies on the underlying shell or agent to prevent command injection characters (e.g., semicolons or backticks).
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 17, 2026, 06:36 PM