The Agent Skills Directory

[DATA_EXFILTRATION]: The skill provides instructions and code to read API credentials from ~/.gooseworks/credentials.json.
The extracted api_key is transmitted via an Authorization header to the API base URL api.gooseworks.ai.
Both the local credential file and the destination API endpoint are resources associated with the skill's author, representing standard platform functionality rather than unauthorized exfiltration.
[INDIRECT_PROMPT_INJECTION]: The skill retrieves untrusted content from Instagram, creating a surface for indirect prompt injection attacks.
Ingestion points: Public profile bios, post captions, and reel descriptions scraped through the Scrape Creators API.
Boundary markers: No explicit delimiters or specialized instructions are provided to help the agent distinguish between skill instructions and retrieved social media text.
Capability inventory: The skill has permissions to read local files (for credentials) and communicate with external web services via curl.
Sanitization: There is no documented mechanism for filtering or sanitizing the content returned from Instagram before it enters the agent's context.

instagram-scraper