data-tools

Pass

Audited by Gen Agent Trust Hub on May 16, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [COMMAND_EXECUTION]: The skill provides extensive patterns for executing data manipulation tools (jq, yq, dasel, mlr, qsv) via the Bash tool. These commands are used to query and edit structured data files as part of the skill's primary functionality.
  • [EXTERNAL_DOWNLOADS]: The skill references documentation and repositories for these tools on well-known platforms like GitHub (e.g., jqlang.github.io, github.com/jqnatividad/qsv). These references are to official and well-known technology services.
  • [PROMPT_INJECTION]: The skill processes external data (JSON, CSV, YAML), creating a surface for indirect prompt injection. However, it mitigates risk by mandating structural parsers instead of text-processing tools.
  • Ingestion points: The skill processes user-supplied data files in various structured formats.
  • Boundary markers: No specific boundary markers are defined in the instructions.
  • Capability inventory: The skill uses Bash to run data manipulation tools and Read/Write tools for file access.
  • Sanitization: The skill relies on the inherent parsing logic of the recommended tools to handle data safely.
Audit Metadata
Risk Level
SAFE
Analyzed
May 16, 2026, 05:13 PM
Security Audit — agent-trust-hub — data-tools