The Agent Skills Directory

[COMMAND_EXECUTION]: Executes system commands to analyze project history and display architectural reports.
Evidence: Uses git log --oneline to identify codebase hot spots and utilizes xdg-open, open, or start to open generated HTML files in the browser (SKILL.md).
[EXTERNAL_DOWNLOADS]: Incorporates third-party resources from well-known content delivery networks into the generated reports.
Evidence: Loads Tailwind CSS from cdn.tailwindcss.com and the Mermaid.js diagramming library from cdn.jsdelivr.net (HTML-REPORT.md).
[PROMPT_INJECTION]: The skill processes untrusted content from the codebase and git history, creating a vulnerability to indirect prompt injection.
Ingestion points: Reads codebase files, CONTEXT.md, Architecture Decision Records (ADRs), and git commit logs (SKILL.md).
Boundary markers: Lacks explicit delimiters or instructions to ignore embedded commands within ingested data.
Capability inventory: The agent can write to the local file system, execute shell commands, and invoke other specialized skills like /grilling or /domain-modeling (SKILL.md).
Sanitization: No sanitization or validation of the content read from the repository is implemented.

improve-codebase-architecture