cloud
Pass
Audited by Gen Agent Trust Hub on May 8, 2026
Risk Level: SAFE
Full Analysis
- [COMMAND_EXECUTION]: The skill executes various commands using
uvxanduv run. This includes launching MCP servers likeawslabs.aws-api-mcp-serverandopensearch-mcp-server-py, and running a local deployment scriptscripts/opensearch_ops.py.- [EXTERNAL_DOWNLOADS]: The skill dynamically downloads and executes packages from public registries and a remote AWS-hosted MCP server (https://knowledge-mcp.global.api.aws) using theuvxtool. These sources are associated with the vendor or recognized cloud services.- [DATA_EXFILTRATION]: The skill requires AWS credentials and OpenSearch master passwords to provision resources. These are handled within the expected scope of AWS and OpenSearch management without signs of exfiltration to unauthorized parties.- [PROMPT_INJECTION]: The skill contains instructional content for the agent to guide users through deployment workflows. It uses trigger word expansion to ensure the skill activates for relevant search-related queries, which is a benign instructional design.
Audit Metadata