Mental Models
v1.0.5Provides decision support by applying Charlie Munger’s mental models to reveal non-obvious insights and shifts in framing for judgment calls and strategy.
⭐ 0· 264·2 current·2 all-time
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
OpenClaw
Benign
high confidencePurpose & Capability
Name/description (latticework thinking / Charlie Munger mental models) align with the bundled SKILL.md and the 20 model documents. The skill is instruction-only and does not request unrelated binaries, credentials, or config paths — all consistent with a framing/analysis advisor.
Instruction Scope
Runtime instructions are focused on when to activate and how to structure outputs; they do not instruct access to system secrets or external endpoints. Two small scope concerns: (1) the skill says 'The lattice reads the user profile before every analysis' and includes a user-profile template, but it is vague about where a filled profile should live and how it will be read; (2) the assets mention an automatically updated 'learnings log' across sessions, but there is no install/persistence mechanism in the package to perform automatic updates. These are documentation gaps that could give the agent broad discretion unless the platform constrains file access.
Install Mechanism
No install spec and no code files that execute; this is the lowest-risk model for installation because nothing is written to disk or downloaded at install time.
Credentials
The skill requires no environment variables, credentials, or config paths. The files are content-only summaries of lenses and templates — nothing in requires.env contradicts the description.
Persistence & Privilege
always:false (normal). The skill can be invoked autonomously by default (platform standard). The only minor privilege question is the implied cross-session persistence ('learnings log updated automatically') despite there being no code/install to implement that — either the platform provides persistence or the statement is misleading. If the platform allows skills to store session state, confirm what storage is used and who can read it.
Assessment
This skill is instruction-only and internally consistent with its stated purpose: it contains a set of explicit mental‑model lenses and activation rules and requests no credentials or installs. Before enabling it: (1) note the author/source is unknown and there is no homepage — limited provenance; (2) clarify how the 'user profile' and the 'learnings log' are expected to be provided and persisted (the package contains a template but no code to read/write a user file across sessions); (3) be aware the agent may be allowed to invoke the skill autonomously by default — if you prefer manual control, restrict invocation in your agent settings; (4) if you are concerned about accidental activation or undesired access to user files, run the skill in a sandboxed agent or require explicit user approval before activation. Overall the package is coherent and low-risk, but confirm persistence/storage semantics and invocation policy to match your privacy preferences.Like a lobster shell, security has layers — review code before you run it.
latestvk97ehj4gmpzq3t670jdv7s0f4182d1xy
License
MIT-0
Free to use, modify, and redistribute. No attribution required.