interdisciplinary-real-world-connection-mapper

Pass

Audited by Gen Agent Trust Hub on May 14, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill is designed to interpolate user-provided content directly into the prompt, creating a surface for indirect instructions.
  • Ingestion points: Data enters the agent context through the real_world_problem and primary_subject fields in SKILL.md.
  • Boundary markers: The prompt lacks explicit delimiters or instructions to the model to ignore any malicious commands embedded within the provided problem description.
  • Capability inventory: No scripts, subprocesses, file operations, or network tools are defined or allowed in the skill's configuration.
  • Sanitization: Input is not sanitized or validated before being placed into the prompt template.
  • [SAFE]: No evidence of malicious code, hidden URLs, credential theft, or persistence mechanisms was found. The skill focuses entirely on academic curriculum design using established educational frameworks.
Audit Metadata
Risk Level
SAFE
Analyzed
May 14, 2026, 03:33 PM
Security Audit — agent-trust-hub — interdisciplinary-real-world-connection-mapper