cspt

Pass

Audited by Gen Agent Trust Hub on Nov 9, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill is instructional and focuses on teaching the agent how to identify and remediate security vulnerabilities in external codebases.
  • [SAFE]: Includes legitimate shell commands for auditing source code, such as using grep to find potential vulnerability patterns in local files.
  • [SAFE]: URL encoding and bypass strategies (e.g., double-encoding slashes or using dot-encoding) are provided as educational examples of common web application security bypasses and do not represent malicious intent within the skill itself.
  • [SAFE]: External links point to a technical blog post regarding CSPT research, used as a reference for the skill's content.
Audit Metadata
Risk Level
SAFE
Analyzed
Nov 9, 2026, 12:00 PM