specialist-specialized-specialized-agent-builder

Pass

Audited by Gen Agent Trust Hub on Aug 15, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection due to the way it handles user-provided data.
  • Ingestion points: Untrusted user input enters the agent context through the {the user's question} placeholder in the Agent tool call within SKILL.md.
  • Boundary markers: The skill lacks any delimiters (like XML tags or triple quotes) or specific guardrail instructions to help the subagent distinguish between administrative instructions and the interpolated user content.
  • Capability inventory: The spawned subagent is granted significant capabilities, including the ability to write files to disk, access a 'Bash' tool for command execution, and spawn further agents via the 'Agent' tool.
  • Sanitization: No filtering or validation mechanisms are implemented to ensure the user input does not contain instructions that could override the subagent's persona or safety rules.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 15, 2026, 08:50 PM
Security Audit — agent-trust-hub — specialist-specialized-specialized-agent-builder