speech-recognition

Pass

Audited by Gen Agent Trust Hub on Sep 11, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill acts as an instructional guide that processes user queries to provide coding assistance and architectural reviews.
  • Ingestion points: User prompts requesting code snippets or implementation reviews, as illustrated in the evals/evals.json file.
  • Boundary markers: The documentation uses standard Markdown headers and triple-backtick code blocks to separate descriptive text from Swift implementation examples.
  • Capability inventory: The skill is composed of static Markdown and JSON metadata; it does not include any scripts, executables, or tool definitions that would grant the agent the ability to perform file system operations, execute shell commands, or make network requests.
  • Sanitization: As an informational resource, the skill provides templates and best practices rather than implementing runtime input filtering.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 11, 2026, 11:14 AM
Security Audit — agent-trust-hub — speech-recognition