The Agent Skills Directory

[DATA_EXFILTRATION]: The skill performs network requests to the external domain api.frevana.com to manage media Pins. This involves reading local files for upload and writing them to local paths during download, which are the primary functions of the skill.
[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted metadata (e.g., titles, source URLs, and collection names) retrieved from the Frevana API, which could contain instructions intended to influence the agent's behavior.
Ingestion points: Data is ingested from external API responses in scripts/pins.py and presented to the agent.
Boundary markers: The skill utilizes structured JSON for API communication and CLI output to help separate data from instructions.
Capability inventory: The skill has the capability to perform network requests via urllib.request and write local files using Path.open in scripts/pins.py.
Sanitization: Content is handled through standard JSON parsing; however, there is no specialized filtering for natural language instructions that might be embedded in API metadata.
[COMMAND_EXECUTION]: The skill's logic is contained within a Python script (scripts/pins.py) that is executed via a Bash wrapper (scripts/pins.sh). The script relies solely on the Python standard library and processes user arguments via argparse.

frevana-pins