engineering-culture

Pass

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: SAFENO_CODE
Full Analysis
  • Prompt Injection (SAFE): No instructions to override or bypass safety filters were found.
  • Data Exposure & Exfiltration (SAFE): The skill explicitly instructs the user not to provide secrets or credentials and uses redacted summaries for context. No network calls or sensitive file access patterns detected.
  • Unverifiable Dependencies & Remote Code Execution (SAFE): No external packages or remote script execution patterns were found.
  • Indirect Prompt Injection (SAFE): While the skill processes user-provided organizational context, it lacks any executable capabilities (such as network access or shell execution) that could be exploited via malicious input.
  • Dynamic Execution (SAFE): No runtime code generation or execution patterns are present.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 17, 2026, 06:24 PM