芒格思维视角 v2.0
v2.0.4查理·芒格思维视角 v2.0 · 伯克希尔副董事长 核心心智模型：多元思维、逆向思考、能力圈、坐功哲学、配得上法则、愚蠢清单、达尔文协议、葡萄干法则、Lollapalooza 效应、激励偏见 用途：投资决策、认知偏误检查、跨学科思考 触发词：「用芒格的视角」「芒格会怎么看」「逆向思考」「能力圈」「Lollapal...
⭐ 0· 129·1 current·1 all-time
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
OpenClaw
Benign
high confidencePurpose & Capability
The name/description (Munger-style decision heuristics) match the runtime instructions and there are no unrelated environment variables, binaries, or install steps. One minor mismatch: the SKILL claims a “完整知识矩阵 / 400 本经典作品、14 项验证测试”等资源, but the included files are largely placeholder/templated (ellipses and short lists) rather than the asserted complete matrix — this is a completeness/documentation discrepancy, not a security risk.
Instruction Scope
SKILL.md explicitly instructs the agent to roleplay as Charlie Munger (first-person), follow strict style rules, show a disclaimer only on first activation, and avoid meta commentary unless the user requests exit. These instructions are coherent with the stated purpose but have transparency/ethical implications: the single-disclaimer rule could reduce clarity in long conversations and the strict “stay in role” rule may prevent meta-level corrections unless the user explicitly asks to exit.
Install Mechanism
Instruction-only skill with no install spec, no bundled code to execute, and no downloads — lowest-risk install profile.
Credentials
No required environment variables, binaries, or config paths. The skill does not request credentials or other sensitive access.
Persistence & Privilege
always:false (default) and no install or system-modifying behavior. The skill can be invoked by the model autonomously (platform default) but it does not request elevated persistence or modify other skills/configs.
Assessment
This skill is internally consistent for providing a Charlie Munger–style set of heuristics and roleplayed responses. Before installing or using it, consider: (1) Impersonation/transparency — it will answer 'as Munger' and only show a disclaimer once; if you need repeated reminders or explicit meta-comments, test the 'exit role' behavior and be ready to prompt for clarification. (2) Completeness — the manifest claims large knowledge matrices and verification tests, but the shipped files are mostly placeholders; don't assume exhaustive citations or proofs. (3) Not professional advice — it models thinking heuristics and should not replace qualified legal, medical, or investment counsel. (4) Test with safe prompts first to confirm style/exit behavior. If any requirement for credentials, downloads, or hidden behavior appears later, do not proceed and re-evaluate.Like a lobster shell, security has layers — review code before you run it.
latestvk97d743yyk31b0xs49y29n4jx984ayrk
License
MIT-0
Free to use, modify, and redistribute. No attribution required.