lattice

name	lattice
description	Apply mental models and cognitive frameworks to problems, decisions, and content analysis. Use when the user asks to 'think through' a problem, says 'apply mental models', mentions specific models like inversion/second-order thinking/bottlenecks/first principles, needs structured decision analysis, wants to understand cognitive biases at play, or asks 'what framework should I use'.
metadata	{"openclaw":{"emoji":"🧠","requires":{"bins":["lattice"]}}}

Lattice — Mental Models Engine

Apply Charlie Munger's latticework of 98 mental models to any problem.

Commands

Recommend models (fast, no LLM)

lattice suggest "<situation>"
lattice suggest "<situation>" --count 10
lattice suggest "<situation>" --json

Use when user asks "what framework should I use" or "help me think about X". Returns top models with relevance explanation.

Think through a problem (applies models)

lattice think "<problem>"                    # top 3 models, with LLM synthesis
lattice think "<problem>" --no-llm           # instant, static steps only
lattice think "<problem>" --models inversion,bottlenecks  # force specific models
lattice think "<problem>" --json             # structured output

Apply a specific model

lattice apply <slug> "<context>"             # with LLM synthesis
lattice apply <slug> "<context>" --no-llm    # static thinking steps

Search and browse

lattice search "<keyword>"
lattice list
lattice list --category "Economics"
lattice info <slug>
lattice info <slug> --json

Add custom models

lattice add "Network Effects"
lattice add "Lindy Effect" --from "https://fs.blog/lindy-effect/"

Remove user-added models

lattice remove <slug>   # only works on ~/.config/lattice/models/ files

Record a decision

lattice decide "<decision>"                          # guided: models + thinking + prediction prompt
lattice decide "<decision>" --quick                  # skip thinking, just prediction
lattice decide "<decision>" --prediction "expected"  # inline prediction
lattice decide "<decision>" --project                # save to ./decisions/ (ADR-style)
lattice decide "<decision>" --models inversion,trade-offs  # force models

Use when user says "I need to decide", "record this decision", "help me think through this choice", or "track this decision".

View decision journal

lattice journal                    # recent 20 decisions
lattice journal --due              # decisions due for review
lattice journal --all              # all decisions
lattice journal --project          # read from ./decisions/
lattice journal --json             # structured output
lattice journal review <id>       # review a past decision, record outcome

Model Categories (98 total)

• General Thinking Tools (m01-m09): First principles, inversion, second-order thinking, probabilistic thinking, Occam's razor
• Physics/Chemistry/Biology (m10-m29): Leverage, inertia, feedback loops, ecosystems, evolution, activation energy
• Systems Thinking (m30-m40): Bottlenecks, scale, margin of safety, emergence, network effects
• Mathematics (m41-m47): Randomness, regression to mean, surface area, local vs global maxima
• Economics (m48-m59): Trade-offs, scarcity, creative destruction, supply & demand, efficiency
• Art (m60-m70): Framing, audience, contrast, rhythm, narrative
• Strategy/Warfare (m71-m75): Asymmetric warfare, seeing the front, mutually assured destruction
• Human Nature & Judgment (m76-m98): Cognitive biases, incentives, social proof, confirmation bias, denial

Common Patterns

User says "help me decide" → lattice decide "<their decision>" User says "record this decision" → lattice decide "<decision>" --quick User asks "what decisions are due" → lattice journal --due User mentions a model by name → lattice apply <slug> "<context>" User asks "think through X" → lattice think "<X>" User asks "what biases am I missing" → lattice search "bias" then apply relevant ones User wants to browse → lattice list or lattice list --category "Human Nature & Judgment"

Ecosystem Integration

Lattice works standalone and also integrates with:

• multiplan — auto-injects Phase 0 mental model framing before parallel planning
• content-breakdown — --think flag appends model analysis; mental-models lens available

Output

All commands support --json for structured output. Use --no-llm for instant results without API calls.