The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from external websites via web_search to extract earnings details, analyst expectations, and options data. This creates a surface where malicious instructions embedded in third-party financial previews could influence the agent's behavior.
Ingestion points: Web search results for ticker-specific earnings dates, whisper numbers, and analyst expectations (Phases 1, 2, and 5).
Boundary markers: Absent. The instructions do not explicitly tell the agent to ignore instructions found within the search results.
Capability inventory: web_search for data gathering, execute_code (Python) for calculating expected moves, and file writing to generate the final report.
Sanitization: Absent. The skill does not describe any validation or filtering of the content retrieved from the web before using it in calculations or report generation.
[COMMAND_EXECUTION]: The skill utilizes the execute_code tool to perform mathematical operations (Phase 5). While the provided template is a standard calculation for implied volatility and expected moves, the use of this capability alongside untrusted web data represents a standard risk factor for agent skills.

trade-earnings