skills/djalmajr/skills/agile-design/Gen Agent Trust Hub

agile-design

Pass

Audited by Gen Agent Trust Hub on Sep 10, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [SAFE]: The skill defines a standard process for design organization and documentation, utilizing stable IDs and paired notes to ensure traceability between planning artifacts and visual designs.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes project planning files and existing design assets to guide the prototyping process, which represents a standard data ingestion surface for this type of workflow.
  • Ingestion points: Reads requirement and story data from the planning/ directory, project-level DESIGN.md files, and existing design tool canvases.
  • Boundary markers: The instructions do not define specific delimiters or "ignore instructions" markers for the ingested data.
  • Capability inventory: The skill has the capability to write to the project repository (creating README.md and notes) and to interact with external design tools via specialized MCP adapters as described in references/tools.md.
  • Sanitization: No explicit validation or sanitization is performed on the ingested planning data before it is used to populate documentation or drive tool actions.
  • [COMMAND_EXECUTION]: The skill orchestrates operations through various tool-specific adapters and MCP extensions (e.g., Paper MCP, Figma skills, Penpot MCP). These are standard platform integrations necessary for interacting with design software and do not involve high-risk command execution.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 10, 2026, 07:06 PM
Security Audit — agent-trust-hub — agile-design