The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes the Bash tool to execute local scripts (python3 _scripts/gemini-character.py) and perform network operations via curl for media processing workflows.
[EXTERNAL_DOWNLOADS]: The skill communicates with the RunPod API (https://api.runpod.ai) to submit video generation jobs and poll for status updates. RunPod is a well-known AI infrastructure service, and the requests use environment variables for authentication.
[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection by ingesting untrusted external data from brand reference files.
Ingestion points: The skill instructions mandate reading style guides and prompt templates from brands/{brand-slug}/brand-ref.md to be used as base prompts.
Boundary markers: None; the content from external files is interpolated directly into generation prompts without delimiters or instructions to ignore embedded commands.
Capability inventory: The skill has access to the Bash tool, enabling file system access, network operations, and script execution.
Sanitization: There is no evidence of validation, escaping, or filtering applied to the external brand reference data before it enters the agent's context.

c-ai-media