ps-principle-prove-it-works

Pass

Audited by Gen Agent Trust Hub on Sep 1, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill consists of purely instructional content intended to guide agent behavior towards rigorous verification of output. It contains no executable code, network exfiltration patterns, or obfuscation.
  • [INDIRECT_PROMPT_INJECTION]: The instructions encourage the agent to read and inspect artifacts (files, git diffs, runtime behavior) to verify tasks. This creates a surface for indirect prompt injection if the ingested data contains malicious instructions. However, this is inherent to the verification process and no specific vulnerable implementations are provided in the skill.
  • [DYNAMIC_EXECUTION]: The skill recommends writing and running verification scripts to automate the check of artifacts. While this involves code generation and execution, it is presented as a developer best practice for verification within a controlled environment.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 1, 2026, 01:28 AM
Security Audit — agent-trust-hub — ps-principle-prove-it-works