skills/mryll/skills/study-session/Gen Agent Trust Hub

study-session

Pass

Audited by Gen Agent Trust Hub on Aug 10, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [SAFE]: The skill provides a structured framework for code comprehension and mental model recovery. It does not invoke external network resources, execute shell commands, or access sensitive system credentials.
  • [PROMPT_INJECTION]: The skill processes untrusted repository data and tracking pages, creating a surface for indirect prompt injection.
  • Ingestion points: Local source code files and the project's comprehension tracking page.
  • Boundary markers: Absent; instructions do not define specific delimiters for separating code content from agent instructions.
  • Capability inventory: Read access to repository files and write access to update tracking documents and session logs.
  • Sanitization: Absent; the agent is instructed to paste user-provided summaries verbatim into the documentation.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 10, 2026, 08:00 PM
Security Audit — agent-trust-hub — study-session