codemagic

Pass

Audited by Gen Agent Trust Hub on Nov 18, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill instructions include the installation of the official Membrane CLI tool (@membranehq/cli) from the standard npm registry. This is a legitimate vendor-provided tool required for the skill to function within the Membrane ecosystem.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes data retrieved from the Codemagic API, which represents a potential surface for indirect prompt injection if external data (such as build logs or repository metadata) contains adversarial instructions.
  • Ingestion points: Data retrieved via membrane action run and membrane request commands (e.g., application details, build artifacts).
  • Boundary markers: None explicitly defined in the instructions to separate external data from the system prompt.
  • Capability inventory: The skill can start/cancel builds, manage application caches, and perform proxy HTTP requests to the Codemagic API.
  • Sanitization: No specific sanitization or filtering logic is provided; the skill relies on the underlying AI agent's safety protocols to handle ingested content.
Audit Metadata
Risk Level
SAFE
Analyzed
Nov 18, 2026, 03:02 PM