skills/giikin/skills/giime-components/Gen Agent Trust Hub

giime-components

Pass

Audited by Gen Agent Trust Hub on Mar 3, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The file references/mcp-setup.md provides a configuration for an external MCP server at https://genapi-giime.giikin.com/mcp. This server is used to fetch component documentation. The domain is owned by the vendor (giikin) and represents a legitimate infrastructure reference.
  • [PROMPT_INJECTION]: The skill establishes a workflow in SKILL.md where the agent is instructed to fetch and interpret documentation from a remote API, creating a surface for indirect prompt injection.
  • Ingestion points: Untrusted content is retrieved through the get-giime-docs-sidebar and get-giime-component-doc tools.
  • Boundary markers: There are no specific instructions or delimiters provided to prevent the agent from executing instructions potentially embedded in the fetched Markdown documentation.
  • Capability inventory: The skill is used for code generation and UI guidance. While it does not include direct command execution or file-system writing, it influences the code the agent writes for the user.
  • Sanitization: No validation or sanitization mechanisms are described for the content returned by the external documentation API.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 3, 2026, 03:08 AM