The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill possesses an attack surface for indirect prompt injection as it retrieves and processes external content from fund news, company profiles, and manager biographies.
Ingestion points: External data is ingested via fund news (references/fund-news.md), fund company-detail (references/fund-company-detail.md), and fund manager-detail (references/fund-manager-detail.md).
Boundary markers: The instructions do not specify the use of delimiters or specific warnings to ignore instructions embedded within the retrieved financial data or news articles.
Capability inventory: The skill allows the agent to execute shell commands via the hithink-finance CLI and supports writing results to local files using the --output flag.
Sanitization: There are no instructions provided to sanitize or escape external text before it is returned to the agent's context, which could allow maliciously crafted financial reports to influence agent behavior.
[COMMAND_EXECUTION]: The skill is centered around the execution of the hithink-finance CLI tool. It provides detailed instructions for passing complex parameters, including JSON-formatted strings for backtesting conditions and indicator selections.
The hithink-finance CLI is identified as a vendor-specific tool belonging to the skill author (hithink-tech).

hithink-finance-fund