The Agent Skills Directory

[Indirect Prompt Injection Surface]: The skill is designed to process external materials such as transcripts, emails, and call recordings provided by the user. This creates a potential surface for indirect prompt injection, where instructions embedded in those materials might influence the agent's logic. (1) Ingestion points: User-provided materials in SKILL.md Phase 0.2. (2) Boundary markers: The current instructions do not specify the use of delimiters to isolate this untrusted content. (3) Capability inventory: The skill utilizes search tools and local file writing. (4) Sanitization: There is no defined process for sanitizing input before processing.
[External Resource Reference]: The skill references established professional platforms like LinkedIn and Crunchbase for research purposes and suggests deployment on services like Netlify and Vercel. These interactions are consistent with the skill's stated purpose.
[Dynamic Client-Side Code]: The skill generates interactive HTML deliverables that include JavaScript for features like calculators and workflow diagrams. This code is contained within the generated file to avoid external script dependencies, though it does include references to resources like Google Fonts for styling.

create-an-asset