exploratory-data-analysis

Pass

Audited by Gen Agent Trust Hub on Jun 14, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill documentation and the scripts/eda_analyzer.py script recommend and utilize a wide range of third-party scientific libraries, including biopython, pysam, rdkit, mdanalysis, nd2reader, and others. These represent external dependencies that are typically fetched from public package registries.
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection (Category 8) because it processes untrusted scientific data files and incorporates their content into reports reviewed by the AI agent. A malicious data file could contain instructions designed to hijack the agent's behavior.
  • Ingestion points: The scripts/eda_analyzer.py script reads user-provided files in various formats (CSV, FASTA, HDF5, etc.).
  • Boundary markers: While the analyzer script wraps data summaries in JSON blocks within the generated report, it lacks explicit warnings to the agent to disregard instructions potentially embedded within the data.
  • Capability inventory: The skill has the ability to read local files, execute a Python analysis script, and write markdown reports to the filesystem.
  • Sanitization: The script parses data using established libraries but does not perform sanitization to filter out natural language instructions that might be present in the data fields.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 14, 2026, 09:22 AM
Security Audit — agent-trust-hub — exploratory-data-analysis