skills/nathanfaucett/agents/rewrite/Gen Agent Trust Hub

rewrite

Pass

Audited by Gen Agent Trust Hub on Apr 8, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [SAFE]: No malicious behavior, obfuscation, or data exfiltration patterns were identified in the instructions or metadata. The skill is an instructional template for legitimate software development tasks.\n- [PROMPT_INJECTION]: The skill processes user-provided content such as codebase scope and test commands, which constitutes an indirect prompt injection surface. This is a common characteristic of AI developer tools that interact with code and execute shell commands for testing purposes.\n
  • Ingestion points: Data enters the agent's context through the scope (files/directories) and tests_ci (command strings) inputs.\n
  • Boundary markers: None explicitly mentioned in the skill instructions.\n
  • Capability inventory: The skill is designed to modify repository files and execute shell commands for integration and smoke testing.\n
  • Sanitization: Not present; the skill follows a standard development workflow without explicit input filtering.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 8, 2026, 11:29 AM