The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to process multiple text prompts for video generation, including the capability to ingest data from external CSV files.
Ingestion points: Data enters the skill context through the prompts argument in submission functions within SKILL.md and through the load_batch_from_csv function in references/csv-batch-input.md.
Boundary markers: While the instructions in SKILL.md advise users to validate content policy compliance, the code implementations lack technical boundary markers or explicit delimiters to prevent the agent from interpreting instructions embedded within the prompt data.
Capability inventory: The skill possesses network capabilities using requests and aiohttp to interact with api.klingai.com, as seen in SKILL.md and references/batch-processor-class.md.
Sanitization: The provided Python snippets do not include logic for sanitizing or escaping the interpolated prompt strings, potentially allowing malicious instructions in the input data to influence the generation process or downstream handling.

klingai-batch-processing