specialist-engineering-embedded-firmware-engineer

Pass

Audited by Gen Agent Trust Hub on Aug 15, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill interpolates user-provided content directly into a sub-agent prompt template without using delimiters or boundary markers. This creates a surface where a malicious user could attempt to override the sub-agent's persona or instructions.
  • Ingestion points: The variable {the user's question} in the Agent call in SKILL.md.
  • Boundary markers: No explicit delimiters (e.g., XML tags or block quotes) are used to separate the system instructions from the user query.
  • Capability inventory: The sub-agent likely possesses capabilities for file system interaction or command execution, as per its role as a firmware engineer.
  • Sanitization: No sanitization or escaping of the input is performed before interpolation.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 15, 2026, 08:50 PM
Security Audit — agent-trust-hub — specialist-engineering-embedded-firmware-engineer