Mirothinker1.7

深度思考与多维分析 AI Skill。通过发散、收敛、批判、创造四种思维模态对复杂问题进行结构化拆解，输出 ASCII 思维导图和 Markdown 分析报告。Triggers: 深度思考, 深度分析, 帮我分析, 帮我想清楚, 多维分析, 结构化思考, 决策分析, 帮我做决策, 我要做决定, 创意分析, 头脑风...

Security Scan

The skill's code and runtime instructions are consistent with its stated purpose (structured deep-thinking assistant); it calls the local OpenClaw agent and uses Python, writes temporary reports to /tmp, and does not request credentials or contact unexpected endpoints.

Purpose & Capability

The skill is a local deep-thinking/analysis tool and its scripts implement the four thinking stages by calling the OpenClaw agent and rendering results. One minor mismatch: the declared 'Required binaries' list is empty, but the scripts clearly require the openclaw CLI and python3 (the README also lists these). Functionally, those binaries are coherent with the described purpose.

Instruction Scope

SKILL.md and the included scripts limit actions to: calling the openclaw agent for LLM responses, parsing those responses, rendering ASCII/Markdown output, and saving reports under /tmp. The instructions do not attempt to read unrelated system files, environment secrets, or post data to external endpoints beyond using the openclaw agent.

Install Mechanism

No install spec is provided (instruction-only from the registry), and all included files are local shell/Python code. Nothing is downloaded or executed from remote URLs during install; risk is limited to running the provided scripts.

Credentials

The skill does not declare or require any environment variables or credentials. The runtime parses LLM outputs and uses /tmp for temporary files and report storage. It does not attempt to access credential files or other services.

Persistence & Privilege

The skill is not always-enabled, does not request elevated platform privileges, and does not modify other skills or global agent config. It writes temporary files and optional reports to /tmp (normal for CLI utilities).

Assessment

This skill appears to be what it says: a local CLI-based deep-thinking assistant. Before installing/running, note: (1) the scripts require the openclaw CLI and python3 — ensure you have the official openclaw client installed and trust it; (2) the skill writes temporary and report files to /tmp and can save Markdown reports there; (3) it does not request API keys or secrets, but the openclaw agent it invokes may itself connect to remote services depending on your openclaw configuration — review your openclaw agent settings/logging if you are concerned about network calls or data retention; (4) because this is runnable code, consider reviewing the provided scripts locally before executing them.

AI 深度思考技能 —— 四维思维模态引擎。

输入任何复杂问题，通过发散 → 收敛 → 批判 → 创造四个阶段，输出结构化思维导图和行动方案。

使用方式

# 直接输入问题
深度分析：我要不要辞职创业？

# 快速模式（跳过批判模态，更快）
帮我快速分析：如何提升用户留存率

# 保存 Markdown 报告
帮我深度思考并保存报告：产品定价策略

思维模态

模态	职责
P0 问题解构	识别问题类型、约束、成功指标
M1 发散	无限展开可能性，生成 6-8 个维度
M2 收敛	提炼核心洞察，构建优先级矩阵
M3 批判	挑战假设，识别盲点和风险
M4 创造	综合输出突破性方案 + 一句话答案