Systems Thinking Analyst
Explains recurring problems via feedback loops, archetypes and leverage points
- Repository
- danielmiessler/LifeOS
- Stars
- 19,682
- Licence
- MIT
- Files
- 9 files
What it does
Once installed, the agent analyses a recurring problem by looking for the underlying structure that keeps producing it, rather than blaming the last event or a person. It can walk from symptoms down to causes (iceberg model), draw feedback loop diagrams, match the situation to known patterns such as shifting the burden or tragedy of the commons, and rank where a change would have the most effect. It is designed mainly for analysing a solo operator's AI setup, but can also work on real-world systems like a city's water supply.
Great for
- People trying to understand why the same problem keeps coming back
- Operators of AI assistant setups who want to find why a metric or habit keeps drifting
- Students or analysts learning systems thinking and want structured diagrams and archetype checks
Skip it if
- Live incidents or outages where the priority is to stabilise first; the skill says incident causal chains belong to a separate RootCauseAnalysis skill
- Situations where a quick factual answer is needed, since the method is deliberately slow and structural and may over-theorise simple problems
Try asking
- “Analyze why our metric keeps drifting the wrong way, using the numbers I'll paste in”
- “Draw a causal loop diagram of how supply, demand and pricing interact for our city's water”
- “Which archetype matches this pattern where every fix makes the problem come back?”
- “Where is the highest-leverage place to intervene in this recurring problem?”
Install in 3 steps
- 1
Download the skill
The ZIP contains the 9 original files, the MIT licence and an attribution note.
SystemsThinking.zip - 2
Add it to your agent
Claude Code (all projects)
mkdir -p ~/.claude/skills && unzip SystemsThinking.zip -d ~/.claude/skills/Claude Code / Agent SDK (this project only)
mkdir -p .claude/skills && unzip SystemsThinking.zip -d .claude/skills/Claude apps (web/desktop): Settings → Capabilities → Skills → Upload skill, and choose the ZIP.
- 3
Ask for the task
No special command needed — the agent reads the skill's description and loads it when your request matches. Try one of the prompts above.
Get better results
- Provide measured facts (counts, rates, dates) in your request so the analysis can be checked against them later
- Treat the output as a structured hypothesis: check each causal link against your own evidence before acting
- Use it for recurring, structural problems, not one-off incidents
What's inside
- Archetypes.md
- Foundation.md
- Intents/CausalLoop.md
- Intents/ConceptMap.md
- Intents/FindArchetype.md
- Intents/FindLeverage.md
- Intents/Iceberg.md
- LeveragePoints.md
- SKILL.md
Redistributed unmodified from danielmiessler/LifeOS under the MIT licence. The assessment was drafted with AI from the skill's files on 2026-10-29. Skills can instruct your agent to run code — read SKILL.md and any scripts before installing.