The Agent Skills Directory

[SAFE]: The skill serves as a standard interface for the Alpha Vantage financial API. All code samples provided for the agent to use are benign and focus on data retrieval and analysis using established libraries.\n- [INDIRECT_PROMPT_INJECTION]: The skill enables the ingestion of untrusted external content from the Alpha Vantage API, such as market news feeds and earnings call transcripts, which could theoretically contain instructions designed to manipulate the agent.\n
Ingestion points: Data is retrieved via the av_get function from endpoints including NEWS_SENTIMENT and EARNINGS_CALL_TRANSCRIPT documented in references/intelligence.md.\n
Boundary markers: The skill instructions do not specify the use of delimiters or provide the agent with guidance on isolating external text from its internal logic.\n
Capability inventory: The skill's capabilities are restricted to network communication with alphavantage.co and local data processing with pandas. It does not have access to sensitive file writing or arbitrary command execution.\n
Sanitization: No explicit content filtering or sanitization logic is present in the provided documentation or scripts.

alpha-vantage