The Agent Skills Directory

[COMMAND_EXECUTION]: The scripts/seederive.py script uses subprocess.check_call to automatically install the requests library if it is missing from the environment.
[EXTERNAL_DOWNLOADS]: The skill fetches the requests package from the official Python Package Index (PyPI) at runtime to facilitate API communication.
[PROMPT_INJECTION]: The skill processes external data (CSV, Excel, JSON) for labeling tasks, which creates a surface for indirect prompt injection. This is a characteristic of the tool's primary purpose for data labeling.
Ingestion points: The scripts/seederive.py script reads user-provided data via arguments like --file, --raw-data, and --raw-data-file across various management tasks.
Boundary markers: There are no explicit markers or instructions used to delimit the untrusted user data from the system prompts during processing.
Capability inventory: The skill is capable of performing network requests to Volcengine's API gateway, reading and writing local files, and executing package installation commands.
Sanitization: No specific sanitization or filtering logic was identified for the content of the data ingested from external files.

byted-data-label