swift-concurrency

Pass

Audited by Gen Agent Trust Hub on Apr 15, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill instructs the agent to fetch additional documentation from 'sosumi.ai', an unofficial community mirror of Apple's developer documentation, when information is not found locally.
  • [PROMPT_INJECTION]: The use of an external documentation source provides a surface for indirect prompt injection.
  • Ingestion points: Markdown content retrieved from 'sosumi.ai' (as directed in SKILL.md).
  • Boundary markers: Absent; fetched content is not delimited or marked as untrusted.
  • Capability inventory: The agent utilizes the content to generate technical explanations and code snippets.
  • Sanitization: Absent; the skill does not verify the integrity or safety of the fetched documentation.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 15, 2026, 01:53 AM