The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes a local script scripts/extract-video-frames.js to process video files. This execution is central to the skill's stated purpose of video-based debugging and is triggered by specific user commands.
[DATA_EXPOSURE]: The skill reads user-provided video files and application source code. This access is required for the tool to perform its primary function of identifying temporal UI glitches and recommending code-level fixes.
[PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface because it processes external video data that could theoretically contain instructions intended to influence the agent's behavior during code modification.
Ingestion points: User-provided video files and extracted frame images are ingested into the agent's context in SKILL.md.
Boundary markers: No explicit delimiters or instructions are used to separate the analyzed video content from the agent's core instructions.
Capability inventory: The skill can execute local scripts (node), read project files, and modify source code (Step 5: Interactive Fix).
Sanitization: There is no evidence of visual or text-based sanitization for the content extracted from the video frames before it is used to guide code changes.

debug-video