check-error-handling

Pass

Audited by Gen Agent Trust Hub on May 13, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill is a legitimate development tool with no evidence of malicious intent or hidden operations.
  • [DATA_EXFILTRATION]: No exfiltration patterns found. Reference code examples interact with local API paths and standard browser APIs.
  • [REMOTE_CODE_EXECUTION]: No remote code execution or dynamic execution patterns were identified in the skill logic.
  • [PROMPT_INJECTION]: The skill has an ingestion surface for indirect prompt injection via the code files it scans, but it is considered safe due to its limited scope and lack of exploitable capabilities. Ingestion points: user-provided source code. Boundary markers: none. Capability inventory: structural analysis and console reporting. Sanitization: none.
Audit Metadata
Risk Level
SAFE
Analyzed
May 13, 2026, 11:55 AM
Security Audit — agent-trust-hub — check-error-handling