paper-to-agent-guide

Pass

Audited by Gen Agent Trust Hub on Apr 30, 2026

Risk Level: SAFENO_CODE
Full Analysis
  • [SAFE]: The skill consists entirely of instructional markdown and does not include any executable code, scripts, or shell commands.
  • [SAFE]: No sensitive file access, network operations, or hardcoded credentials were detected.
  • [SAFE]: No obfuscation techniques or prompt injection attempts were found within the instructions.
  • [SAFE]: Indirect Prompt Injection Surface analysis: 1. Ingestion points: Processes academic research papers (SKILL.md). 2. Boundary markers: None explicitly defined in the parsing workflow. 3. Capability inventory: Performs knowledge extraction and conversational Q&A. 4. Sanitization: Advises the agent to verify claims against the source text.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 30, 2026, 12:17 PM