NYC

dag-context-bridger

Pass

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: SAFENO_CODE
Full Analysis
  • Prompt Injection (SAFE): The skill is focused on context management and does not contain any instructions aimed at overriding agent behavior or leaking system prompts.
  • Data Exposure & Exfiltration (SAFE): No sensitive file paths, hardcoded secrets, or suspicious network operations were detected.
  • Obfuscation (SAFE): The file contains no encoded content, hidden characters, or homoglyphs.
  • Unverifiable Dependencies & Remote Code Execution (SAFE): No external package managers or remote code execution patterns are present.
  • Indirect Prompt Injection (SAFE): Although the skill defines logic for handling external data (context), it lacks executable code or specific vulnerabilities that would allow for prompt injection through processed data.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 17, 2026, 06:04 PM