The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill documentation includes an installation command for a required companion dependency, draw-diagram, using npx skills add singh-gur/agent_skills. This refers to a package from the same author and follows standard installation conventions for this platform.
[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection (Category 8) because it is designed to ingest and process untrusted data from the web.
Ingestion points: External data enters the agent context via the web_search and fetch_content tools used during the 'Discovery and research' phase to gather current information on technologies and pricing.
Boundary markers: The skill does not explicitly instruct the agent to use delimiters or specific markers to isolate content retrieved from external sources from its own instructions.
Capability inventory: The agent has the capability to write files to the local file system (e.g., producing architecture documents at docs/architecture.md).
Sanitization: The skill includes instructions to cite authoritative sources and distinguish between researched facts and assumptions, providing a level of verification for the generated content.

arch-design