The Agent Skills Directory

[SAFE]: The skill provides documentation and guidance for using the brand.yml standard. No malicious code, obfuscation, or high-risk command execution patterns were detected.
[INDIRECT_PROMPT_INJECTION]: The skill defines a workflow where the agent reads and modifies user-provided _brand.yml files and project code. Ingestion points: User-provided _brand.yml files and project source files (R/Python/Quarto). Boundary markers: Not specified in the instructions. Capability inventory: The agent uses standard text processing and file-handling tools; no arbitrary command execution or network exfiltration capabilities are explicitly requested by the skill instructions. Sanitization: The instructions focus on YAML structure and branding logic, without specific guidance on sanitizing embedded URLs or metadata fields against injection.
[EXTERNAL_DOWNLOADS]: The documentation references installation of standard, well-known packages such as bslib, brand.yml (R), and shiny (Python) from official public registries.
[COMMAND_EXECUTION]: The skill provides examples of standard CLI commands for environment setup and font management, such as pip install and quarto typst fonts, intended for user execution.

brand-yml