The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill defines a workflow that ingests data from untrusted external sources (reference websites), creating an inherent attack surface for indirect prompt injection.
Ingestion points: External content is retrieved via tools like analyze_page_structure(reference) and capture_component(selector) as described in SKILL.md.
Boundary markers: The instructions do not specify the use of delimiters or specific instructions to the agent to ignore potentially malicious commands embedded in the reference site's HTML or text.
Capability inventory: The skill utilizes tools like build_gsap_stub for code generation and verify_build for automated browser-based performance testing.
Sanitization: No sanitization or validation of the text content extracted from the reference site is mentioned in the loop steps.
[DYNAMIC_EXECUTION]: The skill involves the automated generation of code stubs, which is a common developer-aid pattern.
The build_gsap_stub tool is used to transform captured animation data (durations, easings, camera paths) into runnable GSAP and ScrollTrigger JavaScript stubs.
This generation is based on templates and captured data intended for development use and is a core feature of the skill's primary purpose.

awwwards-section-loop