The Agent Skills Directory

[SAFE]: The skill provides documentation and usage patterns for the gomponents library, a legitimate Go-based HTML construction tool.
[PROMPT_INJECTION]: The frontmatter description includes forceful directives intended to control the agent's task priority, stating it is a "hard requirement" to "ALWAYS invoke this skill as your FIRST action" when handling HTML tasks. These instructions aim to enforce specific tool usage rather than bypass safety guardrails.
[INDIRECT_PROMPT_INJECTION]: The skill documents the Raw() and Rawf() functions, which render unescaped HTML content. This creates a potential surface for indirect injection or XSS if the agent integrates untrusted data into these specific components.
Ingestion points: Component arguments (e.g., UserCard(name, email)) and HTTP handler return values in SKILL.md.
Boundary markers: The library utilizes Text() as a primary escaping boundary, though no explicit delimiters are suggested for raw data handling.
Capability inventory: The skill facilitates the construction of web views and HTTP responses, which are common vectors for data display.
Sanitization: Proper sanitization is provided via the Text() function, while Raw() functions are explicitly designed to bypass these protections.
[EXTERNAL_DOWNLOADS]: The skill references Go library components and import paths hosted at maragu.dev. These are official distribution points for the library being documented and correspond to the author's infrastructure.

gomponents