aws-graviton-migration

Pass

Audited by Gen Agent Trust Hub on Jul 27, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill is configured to pull the armswdev/arm-mcp:latest image from Docker Hub. This image is provided by a well-known technology organization (Arm) to facilitate migration analysis.
  • [COMMAND_EXECUTION]: The skill uses the docker run command to execute the MCP server. This is a standard method for containerized tool execution.
  • [PROMPT_INJECTION]: The skill possesses an indirect prompt injection surface as it is designed to scan and analyze external, potentially untrusted source code and dependency manifests (e.g., requirements.txt, Dockerfiles).
  • Ingestion points: Source code files (C++, Python, Go, JS, Java), dependency lists, and Dockerfiles.
  • Boundary markers: No explicit delimiters are defined in the instructions for separating untrusted data from the system prompt.
  • Capability inventory: The skill uses the arm-mcp-server for scanning and knowledge_base_search for verification. It also mentions applying code changes.
  • Sanitization: No explicit sanitization or filtering is mentioned; however, the skill includes a manual checkpoint requiring user confirmation before any code modifications are performed, which significantly reduces the risk of automated malicious actions.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 27, 2026, 03:29 AM
Security Audit — agent-trust-hub — aws-graviton-migration