The Agent Skills Directory

File System Interaction: The scripts/rest_request.py script contains functionality to save raw API responses to a local file path. The raw_output_path parameter is user-controllable, which allows for writing data to arbitrary locations on the file system. This capability could be used to overwrite existing files if not properly restricted.
Network Operations: The skill is designed to communicate with the Ensembl REST API. However, the URL construction logic in the script allows absolute URLs to be passed as the path argument, which could enable the skill to make requests to unintended external domains.
Indirect Prompt Injection Surface: This skill processes data from an external source (the Ensembl API). This represents a potential surface for indirect prompt injection, where malicious instructions embedded in the external data could attempt to influence the agent's behavior.
Ingestion points: Data enters the agent's context through the Ensembl REST API via scripts/rest_request.py.
Boundary markers: No explicit boundary markers or instructions to ignore embedded commands are present in the script's output handling.
Capability inventory: The skill possesses file-writing capabilities (save_raw) and network access (requests).
Sanitization: The script performs structural truncation of the response via the _compact function, but it does not perform content-level sanitization or escaping of the retrieved data.

ensembl-skill