The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses the Bash tool to run python tools/video-analyzer.py "$ARGUMENTS". Passing user-controlled input directly into a shell command is a potential injection vector if the agent environment or the script does not properly sanitize the input.
[PROMPT_INJECTION]: The skill represents an indirect prompt injection surface as it ingests and processes untrusted content from external YouTube videos to generate strategy specifications.
Ingestion points: External YouTube transcripts and metadata processed by tools/video-analyzer.py.
Boundary markers: None identified; the skill instructions do not specify the use of delimiters or 'ignore' instructions for the external data.
Capability inventory: The skill has access to Bash for shell execution, Read for file system access, and is authorized to create project files.
Sanitization: There are no explicit validation or sanitization steps mentioned for the content extracted from the videos before it is presented to the agent for implementation.

video