Skills
skills/steadfastasart/geoscience-skills/striplog/Gen Agent Trust Hub

striplog

Pass

Audited by Gen Agent Trust Hub on Mar 8, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The script scripts/create_striplog.py executes file system operations to read CSV data and export images and structured data files.
  • [PROMPT_INJECTION]: The skill processes untrusted text and CSV data, which represents an indirect prompt injection surface.
  • Ingestion points: scripts/create_striplog.py accepts input via the --csv and --text flags.
  • Boundary markers: No explicit boundary markers or instructions to ignore embedded commands are used during data parsing.
  • Capability inventory: The script performs file I/O operations and data visualization using pandas, matplotlib, and striplog.
  • Sanitization: Data is parsed into structured geological objects via the striplog library, which provides implicit validation of the data format but does not strip potentially malicious instructions from free-text fields.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 8, 2026, 07:41 AM