save

Pass

Audited by Gen Agent Trust Hub on Apr 12, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [SAFE]: The skill's core functionality involves reading and writing to a local wiki/ folder. These actions are consistent with its purpose and the provided instructions.
  • [PROMPT_INJECTION]: The skill extracts information from conversation history, presenting a surface for indirect prompt injection. 1. Ingestion points: Current conversation history (SKILL.md). 2. Boundary markers: Absent; no markers distinguish data from instructions in the prompt. 3. Capability inventory: Read, Write, Edit, Glob, and Grep tools are used for file system interaction (SKILL.md). 4. Sanitization: Absent; the skill does not validate or sanitize extracted conversation content before writing it to files.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 12, 2026, 04:51 PM