The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The scripts/bootstrap-runtime.sh script fetches the core runtime from a remote GitHub repository (https://github.com/JuneYaooo/cheap-claude-video.git). This resource belongs to the skill's author.
[REMOTE_CODE_EXECUTION]: The skill facilitates the execution of remote code by cloning a repository and subsequently running its scripts and installing its dependencies via pip install -r lib/requirements.txt.
[COMMAND_EXECUTION]: The skill relies on extensive use of local shell commands to execute Python-based tools for video synthesis, storyboard validation, and quality assurance.
[INDIRECT_PROMPT_INJECTION]: The skill has an attack surface for indirect prompt injection during its media analysis phase.
Ingestion points: The analyze_video_to_capsule.py script parses external video files to create structured capsule drafts.
Boundary markers: The instructions do not specify any delimiters or safety markers to isolate potentially malicious data embedded in processed media.
Capability inventory: The skill can write to the local filesystem (output and capsule directories) and execute a wide range of production scripts as subprocesses.
Sanitization: There is no evidence of sanitization or filtering of the content derived from analyzed videos before it is used to influence the agent's future actions.

capsule-cinema