cognitive-compile

Pass

Audited by Gen Agent Trust Hub on May 8, 2026

Risk Level: SAFENO_CODEPROMPT_INJECTION
Full Analysis
  • [SAFE]: The skill defines a qualitative process for thinking and documentation. It does not perform any automated system operations, network requests, or sensitive file access.
  • [NO_CODE]: The skill is composed strictly of markdown text providing a framework for the AI's reasoning process. There are no scripts (Python, JavaScript, Bash) or external dependencies included.
  • [PROMPT_INJECTION]: The skill is designed to analyze external data sources and generate wiki notes. This ingestion of untrusted data represents an indirect prompt injection surface.
  • Ingestion points: External source or "raw information" (SKILL.md)
  • Boundary markers: None specified in the instructions
  • Capability inventory: Writes formatted analysis results to a wiki and updates a log file (SKILL.md)
  • Sanitization: Not specified; relies on the agent's interpretation and synthesis logic
Audit Metadata
Risk Level
SAFE
Analyzed
May 8, 2026, 07:03 PM