create-attack-technique

Pass

Audited by Gen Agent Trust Hub on Feb 25, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill configuration references official Terraform providers for AWS, Azure, Google Cloud, and Kubernetes from HashiCorp. These are well-known technology providers and are considered safe.
  • [COMMAND_EXECUTION]: The instructions include standard development commands for formatting and documentation, specifically terraform fmt, go fmt, and make docs. These are routine operations within a developer environment.
  • [PROMPT_INJECTION]: The skill acts as a logic engine for reviewing and generating code based on user-provided attack technique descriptions or existing pull requests. While this creates a processing surface for external data, the risk is inherent to the skill's primary purpose as a development and review assistant.
  • Ingestion points: User prompts and provided code/PR content for the Stratus Red Team repository.
  • Boundary markers: No specific delimiters used to separate untrusted code from instructions.
  • Capability inventory: File system write operations for Go and Terraform files, and execution of local formatting binaries.
  • Sanitization: No explicit sanitization or filtering is applied to the input code prior to processing or formatting.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 25, 2026, 10:09 AM