agent-native-reviewer

Pass

Audited by Gen Agent Trust Hub on Jul 10, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: No malicious code, obfuscation, or data exfiltration patterns were detected. The skill consists entirely of high-level instructions for reviewing agent-related workflows.
  • [INDIRECT_PROMPT_INJECTION]: The skill's primary function is to analyze external repository data (code, PRs, docs), which presents a potential surface for indirect prompt injection.
  • Ingestion points: Analyzes SKILL.md files, .agents/ directories, PR feedback, and CLI command outputs within a repository.
  • Boundary markers: None explicitly defined in the instructions to separate untrusted data from the agent's core logic.
  • Capability inventory: Instructs the agent to use Browser, Chrome, Computer Use, and file-editing tools to verify parity.
  • Sanitization: No sanitization of the input text or file content is described.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 10, 2026, 01:19 PM
Security Audit — agent-trust-hub — agent-native-reviewer