paper-to-agent-guide

Pass

Audited by Gen Agent Trust Hub on Sep 2, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and parse external research papers, which constitutes an untrusted data source.
  • Ingestion points: The workflow described in 'Step 1: Structure Extraction' and 'Step 2: Claim Extraction' involves the agent reading and processing the full content of academic papers provided by the user or external sources.
  • Boundary markers: The instructions do not specify the use of delimiters or 'ignore' markers to differentiate between the agent's instructions and the content of the research papers being analyzed.
  • Capability inventory: This skill is purely instructional and does not include any scripts, tools, or binary executables. It refers to external 'Research-Claw' capabilities which are outside the scope of this specific file.
  • Sanitization: There are no explicit instructions for the agent to sanitize or escape content extracted from the papers before presenting it or storing it in the knowledge base.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 2, 2026, 05:36 AM
Security Audit — agent-trust-hub — paper-to-agent-guide