The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to execute several shell commands to install tools and export video files.
npx skills add heygen-com/hyperframes is used to install external capabilities.
npx hyperframes render is used to execute the rendering process.
reel-export is invoked for final output processing.
[EXTERNAL_DOWNLOADS]: The skill fetches and installs remote content from the heygen-com organization.
Evidence: Reference to heygen-com/hyperframes in the Step 3 export instructions.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest untrusted data from external sources (dashboard links or screenshots) to verify milestone numbers, creating a vulnerability surface where instructions embedded in those sources could influence agent behavior.
Ingestion points: SKILL.md (Step 1, item 2: "A screenshot or link to the dashboard that shows it").
Boundary markers: Absent; there are no instructions to the agent to ignore potentially malicious content within the external source.
Capability inventory: The skill has command execution capabilities (npx) and file-write capabilities (generating HTML files).
Sanitization: Absent; the skill does not specify validation or sanitization routines for the data fetched from the dashboard links.
[DYNAMIC_EXECUTION]: The skill utilizes the agent to generate a self-contained HTML file containing JavaScript and Canvas animation logic at runtime based on user-provided parameters.
Evidence: The prompt in Step 2 explicitly asks the agent to "Build one self-contained HTML file (canvas)".

milestone-reveal