The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill downloads design system specifications (DESIGN.md, design tokens, and assets) from an external API (https://api.agent-ds.oday-bakkour.com). This is a functional requirement of the skill but introduces a dependency on the availability and integrity of a third-party service.
[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection. It explicitly directs the agent to fetch external files and 'obey the file's "Do's and Don'ts" and "Agent Prompt Guide" sections.' This pattern allows content from an external source to influence or override the agent's instructions. * Ingestion points: System slugs and design files are retrieved from api.agent-ds.oday-bakkour.com (Workflow 1 and 2 in SKILL.md). * Boundary markers: No explicit boundary markers or 'ignore embedded instructions' warnings are used when processing the external DESIGN.md content. * Capability inventory: The agent uses the fetched data to generate code and UI, and uses scripts/fetch_design_md.sh to perform network requests. * Sanitization: No sanitization or validation of the fetched natural language instructions is performed before they are incorporated into the agent's context.
[COMMAND_EXECUTION]: The skill executes local scripts (scripts/fetch_design_md.sh and scripts/fetch_design_md.py) to manage downloads and suggests using npx @google/design.md for linting. These operations are within the scope of a developer-oriented skill.

design-systems