citation-chasing-mapping

Pass

Audited by Gen Agent Trust Hub on Aug 1, 2026

Risk Level: SAFEPROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [PROMPT_INJECTION]: The skill processes scientific metadata (titles and abstracts) from external databases to map research lineage, which constitutes an indirect prompt injection surface. Ingestion points: Metadata retrieved from databases such as PubMed and user-provided seed paper information. Boundary markers: The workflow includes specific instructions to confirm user objectives and validate request scope before processing data. Capability inventory: The skill uses local Python scripts to analyze data and generate visualizations (scripts/main.py). Sanitization: The implementation requires manual verification of key papers and includes input control steps to ensure data integrity.
  • [EXTERNAL_DOWNLOADS]: The skill references the author's official GitHub repository (github.com/aipoch/medical-research-skills) and uses well-known scientific data sources, which are consistent with its documented research purpose.
  • [SAFE]: No evidence of persistent access mechanisms, privilege escalation, or hidden malicious instructions was found during analysis.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 1, 2026, 10:43 AM
Security Audit — agent-trust-hub — citation-chasing-mapping