The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill ingests and processes untrusted data from multiple external sources, including news articles, SEC filings, and community discussion topics (references/news.md, references/sec-filings.md, references/topic.md). It lacks explicit boundary markers or instructions to treat this content as data rather than instructions. The risk is relevant because the documented toolset includes write capabilities (e.g., longbridge topic create-reply), which could potentially be triggered by malicious instructions embedded in the ingested data.
[COMMAND_EXECUTION]: The skill operates by executing the longbridge CLI tool. It directs users to install the longbridge-terminal package if the command is not available (SKILL.md).
[EXTERNAL_DOWNLOADS]: The skill retrieves financial data, news, and filings from longbridge.com and references information from official regulatory agencies such as the SEC, SFC, and CSRC.

stock-content