The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses shell commands including rg, Select-String, and project-specific Node/PowerShell scripts for transcript processing and validation. It also executes npx hyperframes lint to check project integrity.
[EXTERNAL_DOWNLOADS]: The skill invokes npx hyperframes lint, which fetches and runs code from the npm registry. While central to the skill's purpose, this involves runtime execution of an unversioned package from a remote source.
[REMOTE_CODE_EXECUTION]: The execution of npx hyperframes lint constitutes remote code execution as it dynamically retrieves and runs a tool from a public registry within the local environment.
[INDIRECT_PROMPT_INJECTION]: The skill ingests and acts upon external transcript data, which presents a surface for indirect prompt injection attacks.
Ingestion points: Transcript files and narrations from the project directory are read into the agent context.
Boundary markers: No specific boundary markers or instructions to ignore embedded commands within the transcript content are provided.
Capability inventory: The skill has file-write access for creating beat compositions and updating index.html, alongside command execution capabilities.
Sanitization: There is no explicit requirement for the agent to sanitize or escape transcript text before it is placed into HTML templates or script variables.

hyperframes-video-beats