The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill facilitates the creation of TUI applications that process untrusted data from various sources.
Ingestion points: As detailed in references/components.md, the skill includes various input components (Forms, Text Inputs) and preview components (Markdown, Hex, Text) designed to process user input and external file contents.
Boundary markers: The skill does not provide instructions or patterns for using boundary markers or 'ignore' instructions when data from these components is passed into the agent context.
Capability inventory: The documentation in SKILL.md and references/troubleshooting.md references the use of file system operations such as os.ReadFile and os.Open to load configuration and display file content.
Sanitization: The provided code snippets focus on visual sanitization, such as stripANSI and truncateString in references/emoji-width-fix.md and references/golden-rules.md, rather than security-focused sanitization of external content.

bubbletea