The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill communicates with various external business data platforms (such as Tianyancha, Qichacha, and Aiqicha) and news services (36Kr, Sina Finance, etc.) to collect corporate information. These are well-known and legitimate services required for the skill's primary function.
[COMMAND_EXECUTION]: Scripts within the skill utilize subprocess.run to interact with the local pass password manager for secure credential retrieval and to execute an internal notification script. This use of command execution is restricted to local, defined tasks supporting the skill's workflow.
[PROMPT_INJECTION]: The skill has an attack surface for indirect prompt injection (Category 8). It ingests untrusted news titles and snippets from external websites, which are then processed by an LLM to evaluate financing signals. Maliciously crafted web content could attempt to influence the LLM's conclusions.

healthcare-monitor