skills/jh941213/codex-lattice/eval/Gen Agent Trust Hub

eval

Pass

Audited by Gen Agent Trust Hub on Aug 25, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill serves a legitimate purpose for code evaluation and does not exhibit any malicious behaviors such as data exfiltration, persistence, or obfuscation.
  • [PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection because it ingests untrusted code changes for evaluation.
  • Ingestion points: The skill ingests 'git diff' data, build outputs, and file references through the 'evaluator' sub-agent.
  • Boundary markers: The prompt instructions lack explicit delimiters or instructions to ignore embedded commands within the code being analyzed.
  • Capability inventory: The skill is restricted to a limited toolset including Read, Bash, Grep, and Glob.
  • Sanitization: The skill does not implement sanitization or filtering of the code content it evaluates.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 25, 2026, 01:43 PM
Security Audit — agent-trust-hub — eval