cognitive-fallacies-guard

Pass

Audited by Gen Agent Trust Hub on Sep 17, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [SAFE]: The analyzed skill consists of instructional Markdown files and a JSON rubric. No executable scripts (.py, .js, .sh), binaries, or dangerous system commands were found. The skill operates entirely through natural language instructions to the agent.
  • [INDIRECT_PROMPT_INJECTION]: The skill is designed to process external, untrusted content (visualizations, dashboards, and reports provided by users) to detect fallacies, which introduces a vulnerability surface.
  • Ingestion points: User-provided artifacts such as charts, dashboards, and reports are analyzed according to the workflow in SKILL.md.
  • Boundary markers: No explicit delimiters or instructions are provided to the agent to distinguish user-provided data from system instructions or to ignore embedded commands within the analyzed artifacts.
  • Capability inventory: None. The skill does not define any scripts, subprocess calls, network operations, or file-writing capabilities. It is purely instructional.
  • Sanitization: The skill lacks logic to sanitize or filter user-provided text or data prior to analysis.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 17, 2026, 01:24 PM
Security Audit — agent-trust-hub — cognitive-fallacies-guard