rsn-perceiving-information

Pass

Audited by Gen Agent Trust Hub on Feb 19, 2026

Risk Level: SAFEPROMPT_INJECTIONNO_CODE
Full Analysis
  • [Prompt Injection] (SAFE): No direct malicious instructions, system prompt extraction, or jailbreak attempts were found in the markdown files.
  • [Data Exposure] (SAFE): No hardcoded credentials or sensitive file paths were detected.
  • [Indirect Prompt Injection] (LOW): The skill is designed to process external information from untrusted sources (scanning, monitoring, and focusing modes), which presents a surface for indirect prompt injection. Evidence Chain: 1. Ingestion points: references/scanning.md (web and news sources), references/focusing.md (primary/secondary sources). 2. Boundary markers: Absent. 3. Capability inventory: No executable code provided in the skill. 4. Sanitization: Absent.
  • [No Code] (SAFE): This skill consists entirely of instructional markdown and contains no scripts, binaries, or package dependency files.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 19, 2026, 02:08 PM