The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill functions by ingesting and analyzing data from external, untrusted RSS feeds, creating a surface for potential injection attacks.
Ingestion points: The scripts/sentiment_analyzer.py script fetches content from several external domains (e.g., coindesk.com, cointelegraph.com) using the requests library.
Boundary markers: Absent. The article titles and descriptions are extracted and processed directly without delimiters or instructions to the agent to disregard embedded content.
Capability inventory: The skill includes network fetching capabilities and data processing, but does not exhibit high-risk behaviors like file writing, privilege escalation, or arbitrary code execution.
Sanitization: Absent. There is no evidence of filtering or validation of the fetched article content to prevent the inclusion of malicious instructions.

market-sentiment