teach

Pass

Audited by Gen Agent Trust Hub on Sep 7, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill processes external data like source code and git diffs to generate explanations. This creates a surface where malicious instructions embedded in the analyzed code could potentially influence the agent's behavior or the parameters used when spawning sub-agents.
  • Ingestion points: Reads code, diffs, and subsystem documentation as specified in SKILL.md.
  • Capability inventory: Invokes external skills (how, why), uses spawn_agent for parallel execution, and accesses a debugger or image generation tools.
  • Boundary markers: The instructions do not specify the use of delimiters or warnings to ignore instructions found within the analyzed code.
  • Sanitization: The skill lacks explicit sanitization steps for the data it interpolates into prompts or tool calls.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 7, 2026, 11:13 PM
Security Audit — agent-trust-hub — teach