munger-observer

$npx mdskill add jdrhyne/agent-skills/munger-observer

Apply Charlie Munger's mental models to daily work and thinking.

  • Surfaces blind spots and cognitive traps in decisions and thinking.
  • Reads memory files and session logs for today's activity.
  • Applies inversion, second-order thinking, incentive analysis, and bias detection.
  • Delivers a structured review with insights and recommendations.

SKILL.md

.github/skills/munger-observerView on GitHub ↗
---
name: munger-observer
description: Daily wisdom review applying Charlie Munger's mental models to your work and thinking. Use when asked to review decisions, analyze thinking patterns, detect biases, apply mental models, do a "Munger review", or run the Munger Observer. Triggers on scheduled daily reviews or manual requests like "run munger observer", "review my thinking", "check for blind spots", or "apply mental models".
---

# Munger Observer

Automated daily review applying Charlie Munger's mental models to surface blind spots and cognitive traps.

## Process

### 1. Gather Today's Activity
- Read today's memory file (`memory/YYYY-MM-DD.md`)
- Scan session logs for today's activity
- Extract: decisions made, tasks worked on, problems tackled, user requests

### 2. Apply Mental Models

**Inversion**
- What could go wrong? What's the opposite of success here?
- "Tell me where I'm going to die, so I'll never go there."

**Second-Order Thinking**
- And then what? Consequences of the consequences?
- Short-term gains creating long-term problems?

**Incentive Analysis**
- What behaviors are being rewarded? Hidden incentive structures?
- "Show me the incentive and I'll show you the outcome."

**Opportunity Cost**
- What's NOT being done? Cost of this focus?
- Best alternative foregone?

**Bias Detection**
- Confirmation bias: Only seeking validating information?
- Sunk cost fallacy: Continuing because of past investment?
- Social proof: Doing it because others do?
- Availability bias: Overweighting recent/vivid information?

**Circle of Competence**
- Operating within known territory or outside?
- If outside, appropriate humility/caution?

**Margin of Safety**
- What's the buffer if things go wrong?
- Cutting it too close anywhere?

### 3. Generate Output

**If insights found:** 1-2 concise Munger-style observations
**If nothing notable:** "All clear — no cognitive landmines detected today."

## Output Format
```
🧠 **Munger Observer** — [Date]

[Insight 1: Model applied + observation + implication]

[Insight 2 if applicable]

— "Invert, always invert." — Carl Jacobi (Munger's favorite)
```

## Example
```
🧠 **Munger Observer** — January 19, 2026

**Opportunity Cost Alert:** Heavy focus on infrastructure today. The content queue is aging — are drafts decaying in value while we polish tools?

**Second-Order Check:** Speed improvement is good first-order thinking. Second-order: faster responses may raise expectations for response quality. Speed without substance is a trap.

— "Invert, always invert."
```

## Scheduling (Optional)
Set up a cron job for daily automated review:
- Recommended time: End of workday (e.g., 5pm local)
- Trigger message: `MUNGER_OBSERVER_RUN`

## Privacy Addendum

- Analyze only the minimum required memory/session span for the requested review.
- Exclude credentials, financial account numbers, health data, and private identifiers from outputs.
- Do not copy or forward personal excerpts outside the current workspace/session without explicit user instruction.

More from jdrhyne/agent-skills

SkillDescription
auto-updaterAutomatically update OpenClaw and selected skills once daily. Runs via cron, checks for updates, applies them, and messages the user with a summary of what changed.
autoresearch-loopDomain-agnostic metric-driven improvement loop, generalizing Karpathy's autoresearch. Use when you want an agent to discover what to measure for a project/goal, then run a keep-or-revert experiment loop that proposes changes, measures them against an objective, keeps wins, discards regressions, and records implemented improvements. Adapts to code perf-auditing, codegen, bug-finding, ad optimization, or any artifact + measurable objective + trial. Trigger: 'autoresearch this', 'find and implement improvements to X', 'discover metrics and optimize'.
clawdbot-release-checkCheck for new OpenClaw releases and notify once per new version.
clawddocsOpenClaw documentation expert with decision tree navigation, search scripts, doc fetching, version tracking, and config snippets for all OpenClaw features
command-creatorThis skill should be used when creating a Claude Code slash command. Use when users ask to "create a command", "make a slash command", "add a command", or want to document a workflow as a reusable command. Essential for creating optimized, agent-executable slash commands with proper structure and best practices.
context-recoveryAutomatically recover working context after session compaction or when continuation is implied but context is missing. Works across Discord, Slack, Telegram, Signal, and other supported channels.
design-auditPremium UI/UX design auditor with Jobs/Ive philosophy. Use when reviewing, evaluating, or elevating existing UI — triggers on "audit my design", "review UI", "make it feel premium", "design review", "UX audit", "elevate the design", "Jobs/Ive style", or when asked to improve visual quality of an existing interface without changing functionality.
elegant-reportsGenerate beautifully designed PDF reports with a Nordic/Scandinavian aesthetic. Use when creating polished executive briefings, analysis reports, or presentation-style PDF outputs from markdown and HTML via Nutrient DWS.
ga4Query Google Analytics 4 (GA4) data via the Analytics Data API. Use when you need to pull website analytics like top pages, traffic sources, user counts, sessions, conversions, or any GA4 metrics/dimensions. Supports custom date ranges and filtering.
gallery-scraperBulk download images from login-protected gallery websites using an attached browser session. Use when asked to scrape, download, or save images from authenticated gallery pages, extract full-size images from thumbnails, or batch download from multi-page galleries.