specialist-engineering-compliance-auditor

Pass

Audited by Gen Agent Trust Hub on Aug 15, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill uses a template that interpolates untrusted user input directly into the instructions for a spawned sub-agent, creating a surface for indirect prompt injection.
  • Ingestion points: User-provided input via the {the user's question} variable in SKILL.md.
  • Boundary markers: Absent; there are no delimiters (e.g., XML tags or triple quotes) to separate user content from the agent's instructions.
  • Capability inventory: The skill has the ability to spawn ephemeral sub-agents via the Agent tool.
  • Sanitization: No sanitization or escaping of the user-provided string is performed before interpolation.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 15, 2026, 08:50 PM
Security Audit — agent-trust-hub — specialist-engineering-compliance-auditor