The Agent Skills Directory

[SAFE]: No security issues detected. The skill contains purely instructional content and configuration for an investment decision framework. The provided links to Xueqiu and YouTube are for legitimate research purposes related to the skill's topic.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and analyze external financial, technical, and sentiment data. While this creates an attack surface for indirect prompt injection, the skill lacks dangerous capabilities such as file writing, arbitrary command execution, or automated network exfiltration, rendering the risk negligible.
Ingestion points: Input fields including Asset, sector, market, or macro context and Relevant financial, technical, policy, or sentiment data (SKILL.md).
Boundary markers: None specified in the markdown instructions.
Capability inventory: No executable scripts, network operations, or file system modifications are present in the skill files.
Sanitization: Not implemented within the skill instructions.

fengliu-reverse-weakhand