tailrocks-typescript-refactor

Pass

Audited by Gen Agent Trust Hub on Sep 8, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [SAFE]: The skill implements best practices for security by disabling model invocation (disable-model-invocation: true) and requiring a 'freeze' on behavior before any edits occur.
  • [INDIRECT_PROMPT_INJECTION]: As a code refactoring tool, the skill naturally processes untrusted source files, which is a common vector for indirect prompt injection. However, it mitigates this risk through a mandatory 'Assume untrusted' policy in runtime-trust.md, requiring all data at trust boundaries to be parsed with runtime schemas rather than raw type assertions.
  • [SAFE]: The confidentiality-and-authority.md reference provides defensive instructions for handling secrets, PII, and production credentials, ensuring that the agent does not exfiltrate or reproduce sensitive information discovered during the refactoring process.
  • [SAFE]: No obfuscation, remote code execution, or unauthorized network operations were detected. The skill focuses entirely on localized code transformation and architectural review.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 8, 2026, 03:12 PM
Security Audit — agent-trust-hub — tailrocks-typescript-refactor