Customization
Before executing, check for user customizations at:
${PAI_USER_DIR}/SKILLCUSTOMIZATIONS/Science/
If this directory exists, load and apply any PREFERENCES.md, configurations, or resources found there. These override default behavior. If the directory does not exist, proceed with skill defaults.
- Output text notification:
Running the **WorkflowName** workflow in the **Science** skill to ACTION...
This is not optional. Execute this curl command immediately upon skill invocation.
Science - The Universal Algorithm
The scientific method applied to everything. The meta-skill that governs all other skills.
The Universal Cycle
GOAL -----> What does success look like?
|
OBSERVE --> What is the current state?
|
HYPOTHESIZE -> What might work? (Generate MULTIPLE)
|
EXPERIMENT -> Design and run the test
|
MEASURE --> What happened? (Data collection)
|
ANALYZE --> How does it compare to the goal?
|
ITERATE --> Adjust hypothesis and repeat
|
+------> Back to HYPOTHESIZE
The goal is CRITICAL. Without clear success criteria, you cannot judge results.
Workflow Routing
Output when executing: Running the **WorkflowName** workflow in the **Science** skill to ACTION...
Core Workflows
| Trigger |
Workflow |
| "define the goal", "what are we trying to achieve" |
Workflows/DefineGoal.md |
| "what might work", "ideas", "hypotheses" |
Workflows/GenerateHypotheses.md |
| "how do we test", "experiment design" |
Workflows/DesignExperiment.md |
| "what happened", "measure", "results" |
Workflows/MeasureResults.md |
| "analyze", "compare to goal" |
Workflows/AnalyzeResults.md |
| "iterate", "try again", "next cycle" |
Workflows/Iterate.md |
| Full structured cycle |
Workflows/FullCycle.md |
Diagnostic Workflows
| Trigger |
Workflow |
| Quick debugging (15-min rule) |
Workflows/QuickDiagnosis.md |
| Complex investigation |
Workflows/StructuredInvestigation.md |
Resource Index
| Resource |
Description |
METHODOLOGY.md |
Deep dive into each phase |
Protocol.md |
How skills implement Science |
Templates.md |
Goal, Hypothesis, Experiment, Results templates |
Examples.md |
Worked examples across scales |
Domain Applications
| Domain |
Manifestation |
Related Skill |
| Coding |
TDD (Red-Green-Refactor) |
Development |
| Products |
MVP -> Measure -> Iterate |
Development |
| Research |
Question -> Study -> Analyze |
Research |
| Prompts |
Prompt -> Eval -> Iterate |
Evals |
| Decisions |
Options -> Council -> Choose |
Council |
Scale of Application
| Level |
Cycle Time |
Example |
| Micro |
Minutes |
TDD: test, code, refactor |
| Meso |
Hours-Days |
Feature: spec, implement, validate |
| Macro |
Weeks-Months |
Product: MVP, launch, measure PMF |
Integration Points
| Phase |
Skills to Invoke |
| Goal |
Council for validation |
| Observe |
Research for context |
| Hypothesize |
Council for ideas, RedTeam for stress-test |
| Experiment |
Development (Worktrees) for parallel tests |
| Measure |
Evals for structured measurement |
| Analyze |
Council for multi-perspective analysis |
Key Principles (Quick Reference)
- Goal-First - Define success before starting
- Hypothesis Plurality - NEVER just one idea (minimum 3)
- Minimum Viable Experiments - Smallest test that teaches
- Falsifiability - Experiments must be able to fail
- Measure What Matters - Only goal-relevant data
- Honest Analysis - Compare to goal, not expectations
- Rapid Iteration - Cycle speed > perfect experiments
Anti-Patterns
| Bad |
Good |
| "Make it better" |
"Reduce load time from 3s to 1s" |
| "I think X will work" |
"Here are 3 approaches: X, Y, Z" |
| "Prove I'm right" |
"Design test that could disprove" |
| "Pretend failure didn't happen" |
"What did we learn?" |
| "Keep experimenting forever" |
"Ship and learn from production" |
Quick Start
- Goal - What does success look like?
- Observe - What do we know?
- Hypothesize - At least 3 ideas
- Experiment - Minimum viable tests
- Measure - Collect goal-relevant data
- Analyze - Compare to success criteria
- Iterate - Adjust and repeat
The answer emerges from the cycle, not from guessing.
1---2name: science3description: Hypothesis-test-analyze cycles for systematic problem-solving — the meta-skill governing all others. Includes define goal, generate hypotheses, design experiment, measure results, analyze results, iterate, full cycle, quick diagnosis, and structured investigation. USE WHEN think about, figure out, try approaches, experiment with, iterate on, improve, optimize, define goal, generate hypotheses, design experiment, measure results, analyze results, full cycle, quick diagnosis, structured investigation, science, hypothesis.4---56## Customization78**Before executing, check for user customizations at:**9`${PAI_USER_DIR}/SKILLCUSTOMIZATIONS/Science/`1011If this directory exists, load and apply any PREFERENCES.md, configurations, or resources found there. These override default behavior. If the directory does not exist, proceed with skill defaults.121314<!-- ## 🚨 MANDATORY: Voice Notification (REQUIRED BEFORE ANY ACTION)1516**You MUST send this notification BEFORE doing anything else when this skill is invoked.**17181. **Send voice notification**:19 ```bash20 curl -s -X POST http://localhost:8888/notify \21 -H "Content-Type: application/json" \22 -d '{"message": "Running the WORKFLOWNAME workflow in the Science skill to ACTION"}' \23 > /dev/null 2>&1 &24 ```25-->26272. **Output text notification**:28 ```29 Running the **WorkflowName** workflow in the **Science** skill to ACTION...30 ```3132**This is not optional. Execute this curl command immediately upon skill invocation.**3334# Science - The Universal Algorithm3536**The scientific method applied to everything. The meta-skill that governs all other skills.**3738## The Universal Cycle3940```41GOAL -----> What does success look like?42 |43OBSERVE --> What is the current state?44 |45HYPOTHESIZE -> What might work? (Generate MULTIPLE)46 |47EXPERIMENT -> Design and run the test48 |49MEASURE --> What happened? (Data collection)50 |51ANALYZE --> How does it compare to the goal?52 |53ITERATE --> Adjust hypothesis and repeat54 |55 +------> Back to HYPOTHESIZE56```5758**The goal is CRITICAL.** Without clear success criteria, you cannot judge results.5960---616263## Workflow Routing6465**Output when executing:** `Running the **WorkflowName** workflow in the **Science** skill to ACTION...`6667### Core Workflows6869| Trigger | Workflow |70|---------|----------|71| "define the goal", "what are we trying to achieve" | `Workflows/DefineGoal.md` |72| "what might work", "ideas", "hypotheses" | `Workflows/GenerateHypotheses.md` |73| "how do we test", "experiment design" | `Workflows/DesignExperiment.md` |74| "what happened", "measure", "results" | `Workflows/MeasureResults.md` |75| "analyze", "compare to goal" | `Workflows/AnalyzeResults.md` |76| "iterate", "try again", "next cycle" | `Workflows/Iterate.md` |77| Full structured cycle | `Workflows/FullCycle.md` |7879### Diagnostic Workflows8081| Trigger | Workflow |82|---------|----------|83| Quick debugging (15-min rule) | `Workflows/QuickDiagnosis.md` |84| Complex investigation | `Workflows/StructuredInvestigation.md` |8586---8788## Resource Index8990| Resource | Description |91|----------|-------------|92| `METHODOLOGY.md` | Deep dive into each phase |93| `Protocol.md` | How skills implement Science |94| `Templates.md` | Goal, Hypothesis, Experiment, Results templates |95| `Examples.md` | Worked examples across scales |9697---9899## Domain Applications100101| Domain | Manifestation | Related Skill |102|--------|---------------|---------------|103| **Coding** | TDD (Red-Green-Refactor) | Development |104| **Products** | MVP -> Measure -> Iterate | Development |105| **Research** | Question -> Study -> Analyze | Research |106| **Prompts** | Prompt -> Eval -> Iterate | Evals |107| **Decisions** | Options -> Council -> Choose | Council |108109---110111## Scale of Application112113| Level | Cycle Time | Example |114|-------|-----------|---------|115| **Micro** | Minutes | TDD: test, code, refactor |116| **Meso** | Hours-Days | Feature: spec, implement, validate |117| **Macro** | Weeks-Months | Product: MVP, launch, measure PMF |118119---120121## Integration Points122123| Phase | Skills to Invoke |124|-------|-----------------|125| **Goal** | Council for validation |126| **Observe** | Research for context |127| **Hypothesize** | Council for ideas, RedTeam for stress-test |128| **Experiment** | Development (Worktrees) for parallel tests |129| **Measure** | Evals for structured measurement |130| **Analyze** | Council for multi-perspective analysis |131132---133134## Key Principles (Quick Reference)1351361. **Goal-First** - Define success before starting1372. **Hypothesis Plurality** - NEVER just one idea (minimum 3)1383. **Minimum Viable Experiments** - Smallest test that teaches1394. **Falsifiability** - Experiments must be able to fail1405. **Measure What Matters** - Only goal-relevant data1416. **Honest Analysis** - Compare to goal, not expectations1427. **Rapid Iteration** - Cycle speed > perfect experiments143144---145146## Anti-Patterns147148| Bad | Good |149|-----|------|150| "Make it better" | "Reduce load time from 3s to 1s" |151| "I think X will work" | "Here are 3 approaches: X, Y, Z" |152| "Prove I'm right" | "Design test that could disprove" |153| "Pretend failure didn't happen" | "What did we learn?" |154| "Keep experimenting forever" | "Ship and learn from production" |155156---157158## Quick Start1591601. **Goal** - What does success look like?1612. **Observe** - What do we know?1623. **Hypothesize** - At least 3 ideas1634. **Experiment** - Minimum viable tests1645. **Measure** - Collect goal-relevant data1656. **Analyze** - Compare to success criteria1667. **Iterate** - Adjust and repeat167168**The answer emerges from the cycle, not from guessing.**
mj-deving