The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes basic shell commands within its lifecycle hooks for informational purposes and status reporting. Evidence: The pre_execution and post_execution hooks in SKILL.md use standard find, grep, and echo commands to visualize the project structure and verify document creation.
[PROMPT_INJECTION]: The skill has an indirect prompt injection surface because it is designed to read and process external markdown documentation which could contain embedded instructions. 1. Ingestion points: Reads .md and README.md files from the project directory. 2. Boundary markers: No explicit delimiters or instructions are provided to the agent to ignore instructions within the project files. 3. Capability inventory: The agent has Write permissions to specific documentation directories. 4. Sanitization: No content validation or sanitization is performed on the ingested markdown text.
[SAFE]: A minor discrepancy exists in the metadata where the author is listed as 'Claude Code' while the context identifies the vendor as 'ruvnet'; however, this is consistent with a template artifact and does not indicate malicious deception.

agent-arch-system-design