The Agent Skills Directory

[SAFE]: The skill is a collection of natural language instructions designed to guide an AI agent through a 'First Principles' reasoning process. It does not contain executable code, remote dependencies, or network operations.
[INDIRECT_PROMPT_INJECTION]: The skill includes a surface for indirect prompt injection as it ingests user-provided descriptions of problems to be analyzed. However, the risk is negligible as the skill's primary function is internal reasoning and it does not possess dangerous capabilities like writing to the file system or making arbitrary network requests.
Ingestion points: User input is interpolated into the 【填写你的问题】 placeholder in SKILL.md.
Boundary markers: None present; the user input is directly inserted into the prompt template.
Capability inventory: The instructions mention using other agent tools (e.g., minimum-experiment, carl-idea-king), which is standard agent orchestration.
Sanitization: None specified.

first-principles