conventions

Pass

Audited by Gen Agent Trust Hub on Jun 23, 2026

Risk Level: SAFE
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The documentation describes installation of the @anthropic-ai/claude-code package from the NPM registry, which is a tool from a well-known service.
  • [CREDENTIALS_UNSAFE]: Templates utilize standard GitHub Secrets syntax (${{ secrets.ANTHROPIC_API_KEY }}) for secure credential management in environment variables.
  • [PROMPT_INJECTION]: The skill includes explicit security constraints in its prompt templates to ensure agents treat issue content as user data rather than instructions, a proactive measure against indirect prompt injection.
  • [DATA_EXFILTRATION]: Workflow examples process standard GitHub event metadata (issue titles and bodies) for the intended purpose of automated triage and classification.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 23, 2026, 03:52 PM
Security Audit — agent-trust-hub — conventions