Examples

Walkthrough of example scripts demonstrating CAL features and real-world usage.

Overview

The runtime includes four example scripts that progressively demonstrate CAL features, from basic queries to complete closed-loop analysis pipelines.

Example Scripts:

simple-query.cal - Basic entity querying
drift-analysis.cal - Gap measurement with DRIFT
tailwind-cascade.cal - Multi-dimensional cascade analysis
closed-loop-pipeline.cal - Complete 5-layer intelligence pipeline

All examples use the same sample dataset: examples/data/entities.json

Running Examples

bash

# Clone or install
npm install @stratiqx/cal-runtime

# Run any example
cal run examples/simple-query.cal --data examples/data/entities.json

# Verbose output
cal run examples/tailwind-cascade.cal --data examples/data/entities.json --verbose

# Save results
cal run examples/closed-loop-pipeline.cal \
  --data examples/data/entities.json \
  --output results.json

Sample Dataset

All examples use examples/data/entities.json containing four entities:

Entity 1: Tailwind CSS

High-impact cascade example (80% revenue decline, 75% layoffs):

json

{
  "id": "tailwind-css",
  "name": "Tailwind CSS",
  "sound": 9,
  "space": 8,
  "time": 9,
  "impact": "high",
  "baseCost": 300000,
  "dimensions": {
    "D1": { "sound": 5, "space": 6, "time": 7 },
    "D2": { "sound": 9, "space": 8, "time": 8 },
    "D3": { "sound": 9, "space": 9, "time": 8 },
    "D5": { "sound": 6, "space": 7, "time": 5 },
    "D6": { "sound": 9, "space": 8, "time": 8 }
  }
}

Entity 2: Stack Overflow

High-impact with traffic decline (35%):

json

{
  "id": "stack-overflow",
  "name": "Stack Overflow",
  "sound": 7,
  "space": 7,
  "time": 7,
  "impact": "high",
  "baseCost": 500000
}

Entity 3: GitHub Copilot

High-impact growth enabler:

json

{
  "id": "github-copilot",
  "name": "GitHub Copilot",
  "sound": 8,
  "space": 9,
  "time": 8,
  "impact": "high",
  "baseCost": 0
}

Entity 4: Stable Enterprise Co

Low-impact baseline:

json

{
  "id": "stable-company",
  "name": "Stable Enterprise Co",
  "sound": 3,
  "space": 4,
  "time": 3,
  "impact": "low",
  "baseCost": 100000
}

Example 1: Simple Query

File: examples/simple-query.cal

Basic entity querying with filters.

Script

cal

-- ================================================
-- Cormorant Agentic Language (CAL)
-- Example: Simple Entity Query
-- ================================================

-- Find all high-impact entities
FORAGE entities
WHERE impact = "high"
SURFACE results

Concepts Demonstrated

FORAGE keyword for querying
WHERE clause filtering
SURFACE for output

Running

bash

cal run examples/simple-query.cal --data examples/data/entities.json

Expected Output

📦 results:
   Type: Array
   Count: 3

The query returns three entities with impact = "high":

Tailwind CSS
Stack Overflow
GitHub Copilot

Key Takeaway

FORAGE + WHERE + SURFACE is the basic pattern for querying and filtering entities.

Example 2: DRIFT Analysis

File: examples/drift-analysis.cal

Gap measurement using the DRIFT formula.

Script

cal

-- ================================================
-- Cormorant Agentic Language (CAL)
-- Example: DRIFT Gap Analysis
-- ================================================

-- Find entities for analysis
FORAGE entities
WHERE sound > 5
SURFACE target_entities

-- Measure methodology-performance gap
DRIFT target_entities
METHODOLOGY 85
PERFORMANCE 40

-- Alert if gap is significant
CHIRP warning "Significant DRIFT detected in target entities"

Concepts Demonstrated

DRIFT keyword for gap measurement
METHODOLOGY and PERFORMANCE values (0-100 scale)
CHIRP for alerting
Chaining multiple actions

Formula

DRIFT = Methodology - Performance
DRIFT = 85 - 40 = 45 (teaching gap)

A positive DRIFT indicates a teaching gap: methodology exceeds performance, requiring more explanation.

Running

bash

cal run examples/drift-analysis.cal --data examples/data/entities.json

Expected Output

⚠️ ═══════════════════════════════════════════════
   CHIRP: WARNING
   TIME: 2026-01-14T10:30:00.000Z
   ───────────────────────────────────────────────
   Significant DRIFT detected in target entities
   ═══════════════════════════════════════════════

Key Takeaway

DRIFT measures gaps between expected (methodology) and actual (performance) states for adaptive behavior.

Example 3: Tailwind Cascade

File: examples/tailwind-cascade.cal

Real-world cascade analysis based on the Tailwind CSS case study.

Script

cal

-- ================================================
-- Cormorant Agentic Language (CAL)
-- Example: Tailwind CSS Cascade Analysis
-- ================================================

-- Find affected entities with high urgency signals
FORAGE entities
WHERE sound > 7
AND impact = "high"
ACROSS D1, D2, D3, D5, D6
DEPTH 3
SURFACE cascade_map

-- Deep dive into revenue dimension
DIVE INTO revenue
WHEN decline > 80
TRACE cascade
EMIT financial_impact

-- Monitor for ongoing signals
PERCH ON segment:"open-source"
LISTEN FOR disruption signals, churn signals
WAKE AFTER 30d
CHIRP warning

-- Output final analysis
SURFACE report AS json

Concepts Demonstrated

ACROSS - Multi-dimensional analysis (D1, D2, D3, D5, D6)
DEPTH - Cascade depth control (3 levels)
DIVE INTO - Deep analysis with conditions
TRACE cascade - Follow cascade pathways
PERCH ON - Observation positioning
LISTEN FOR - Signal monitoring
WAKE AFTER - Time-based triggers
Complex multi-action workflows

Dimensions Analyzed

D1 (Customer) - Customer trust damage
D2 (Employee) - 75% layoffs, knowledge loss
D3 (Revenue) - 80% revenue decline
D5 (Quality) - Skeleton crew for 75M downloads
D6 (Operational) - AI broke discovery funnel

Running

bash

cal run examples/tailwind-cascade.cal \
  --data examples/data/entities.json \
  --verbose

Expected Results

The analysis reveals:

3 entities with high urgency (sound > 7)
Multiple dimensions affected across each entity
Cascade pathways showing how impacts spread
Financial multipliers based on dimensions × depth
Watchers set up for ongoing monitoring

Key Takeaway

CAL can model complex real-world cascades with multi-dimensional analysis, deep dives, and continuous monitoring.

Example 4: Closed-Loop Pipeline

File: examples/closed-loop-pipeline.cal

Complete 5-layer intelligence pipeline: Sense → Analyze → Measure → Decide → Act.

Script

cal

-- ================================================
-- Cormorant Agentic Language (CAL)
-- Example: Full Closed-Loop Intelligence Pipeline
--
-- Sense → Analyze → Measure → Decide → Act
-- ================================================

-- LAYER 1: SENSE
-- Find affected entities with high urgency signals
FORAGE entities
WHERE sound > 7
AND impact = "high"
ACROSS D1, D2, D3, D5, D6
DEPTH 3
SURFACE cascade_map

-- LAYER 2: ANALYZE
-- Deep dive into the cascade
DIVE INTO revenue
WHEN decline > 50
TRACE cascade
EMIT impact_analysis

-- LAYER 3: MEASURE (DRIFT)
-- Measure the gap between methodology and performance
DRIFT cascade_map
METHODOLOGY 85
PERFORMANCE 35

-- LAYER 4: DECIDE (FETCH)
-- Determine action based on Fetch score
-- Fetch = Chirp × |DRIFT| × Confidence
FETCH cascade_map
THRESHOLD 1000
ON EXECUTE CHIRP critical "Immediate action required - cascade in progress"
ON CONFIRM CHIRP warning "Review recommended before action"
ON QUEUE SURFACE queue_report
ON WAIT PERCH ON segment:"monitor"

-- LAYER 5: ACT
-- Output final results
SURFACE results AS json

The 5-Layer Intelligence Model

Layer 1: SENSE

Detect signals in the environment.

cal

FORAGE entities
WHERE sound > 7
ACROSS D1, D2, D3, D5, D6
DEPTH 3

Query entities with high urgency (sound > 7)
Analyze across 5 dimensions
Trace 3 levels deep

Layer 2: ANALYZE

Deep analysis of detected signals.

cal

DIVE INTO revenue
WHEN decline > 50
TRACE cascade

Focus on specific dimensions
Apply conditions (decline > 50)
Map cascade pathways

Layer 3: MEASURE

Quantify gaps and drift.

cal

DRIFT cascade_map
METHODOLOGY 85
PERFORMANCE 35

Calculate DRIFT: 85 - 35 = 50
Large teaching gap indicates significant underperformance
Triggers need for action

Layer 4: DECIDE

Determine appropriate action level.

cal

FETCH cascade_map
THRESHOLD 1000
ON EXECUTE CHIRP critical
ON CONFIRM CHIRP warning
ON QUEUE SURFACE queue_report
ON WAIT PERCH ON segment

FETCH formula:

Fetch = Chirp × |DRIFT| × Confidence

Decision levels:

EXECUTE (>1000): Immediate action
CONFIRM (>500): Review needed
QUEUE (>100): Schedule later
WAIT (≤100): Continue monitoring

Layer 5: ACT

Execute the decided action.

cal

SURFACE results AS json

Output results
Send alerts
Set up monitoring
Queue tasks

Running

bash

cal run examples/closed-loop-pipeline.cal \
  --data examples/data/entities.json \
  --output results.json

Expected Flow

SENSE: 3 high-impact entities detected
ANALYZE: Revenue dimension shows 80% decline
MEASURE: DRIFT = 50 (large teaching gap)
DECIDE: FETCH score > 1000 → EXECUTE level
ACT: Critical alert sent, results output

Key Takeaway

The closed-loop pipeline demonstrates complete intelligence workflows: sensing signals, analyzing patterns, measuring gaps, making decisions, and taking actions.

Common Patterns

Pattern 1: Query + Filter

cal

FORAGE entities
WHERE sound > 7 AND type = "customer"
SURFACE high_priority

Use for: Finding specific entities

Pattern 2: Query + DRIFT

cal

FORAGE targets
WHERE impact = "high"
SURFACE analysis

DRIFT analysis
METHODOLOGY 85
PERFORMANCE 40

Use for: Gap measurement

Pattern 3: Query + FETCH

cal

FORAGE entities
WHERE sound > 7
SURFACE targets

FETCH targets
THRESHOLD 1000
ON EXECUTE CHIRP critical "Act now"

Use for: Decision logic

Pattern 4: Query + Monitor

cal

FORAGE entities
WHERE segment = "enterprise"
SURFACE targets

PERCH ON segment:"enterprise"
LISTEN FOR churn signals
WAKE AFTER 30d
CHIRP warning

Use for: Continuous monitoring

Pattern 5: Full Pipeline

cal

FORAGE entities WHERE sound > 7 SURFACE targets
DRIFT targets METHODOLOGY 85 PERFORMANCE 40
FETCH targets THRESHOLD 1000
  ON EXECUTE CHIRP critical "Act"
SURFACE results

Use for: Complete analysis workflows

Modifying Examples

Adjust Thresholds

cal

-- Original
WHERE sound > 7

-- More sensitive
WHERE sound > 5

-- Less sensitive
WHERE sound > 8

Change Dimensions

cal

-- Original
ACROSS D1, D2, D3, D5, D6

-- Focus on financial
ACROSS D3

-- Include all
ACROSS D1, D2, D3, D4, D5, D6

Adjust Cascade Depth

cal

-- Original
DEPTH 3

-- Shallow analysis
DEPTH 1

-- Deep analysis
DEPTH 5

Modify DRIFT Values

cal

-- Original
METHODOLOGY 85
PERFORMANCE 40

-- Different gap
METHODOLOGY 90
PERFORMANCE 70

Change FETCH Threshold

cal

-- Original
THRESHOLD 1000

-- More aggressive
THRESHOLD 500

-- More conservative
THRESHOLD 2000

Creating Custom Examples

Step 1: Start Simple

cal

-- my-analysis.cal
FORAGE entities
WHERE sound > 6
SURFACE results

Step 2: Add Filtering

cal

FORAGE entities
WHERE sound > 6
AND type = "customer"
AND segment = "enterprise"
SURFACE high_priority

Step 3: Add Analysis

cal

FORAGE entities
WHERE sound > 6
ACROSS D1, D2, D3
DEPTH 2
SURFACE cascade_analysis

Step 4: Add Measurement

cal

FORAGE entities WHERE sound > 6 SURFACE targets
DRIFT targets METHODOLOGY 80 PERFORMANCE 50

Step 5: Add Decision Logic

cal

FORAGE entities WHERE sound > 6 SURFACE targets
DRIFT targets METHODOLOGY 80 PERFORMANCE 50
FETCH targets THRESHOLD 1000
  ON EXECUTE CHIRP critical "Action required"

Step 6: Add Output

cal

FORAGE entities WHERE sound > 6 SURFACE targets
DRIFT targets METHODOLOGY 80 PERFORMANCE 50
FETCH targets THRESHOLD 1000
  ON EXECUTE CHIRP critical "Action required"
SURFACE final_report AS json

Testing Examples

Validate Syntax

bash

cal validate examples/simple-query.cal

Dry Run (Analyze)

bash

cal analyze examples/tailwind-cascade.cal --verbose

Run with Sample Data

bash

cal run examples/simple-query.cal --data examples/data/entities.json

Run with Custom Data

bash

# Create your own data
cat > my-entities.json <<EOF
{
  "entities": [
    { "id": "e1", "name": "Entity 1", "sound": 8, "space": 7, "time": 9 }
  ]
}
EOF

# Run example
cal run examples/simple-query.cal --data my-entities.json

Next Steps

For Learning

Start with simple-query.cal
Progress to drift-analysis.cal
Study tailwind-cascade.cal
Understand closed-loop-pipeline.cal

For Development

Modify examples for your use case
Create custom scripts
Build on the patterns shown
Combine multiple techniques

For Production

Adapt examples to real data
Add error handling
Configure alerts
Set up monitoring

Additional Resources

Getting Started - CAL basics
CLI Reference - Command options
Configuration - Project setup
Data Adapters - Connect your data
Alert Adapters - Configure alerts
Validation - Ensure data quality

Example Repository

All examples are available in the runtime package:

bash

# View examples
ls node_modules/@stratiqx/cal-runtime/examples/

# Copy to your project
cp -r node_modules/@stratiqx/cal-runtime/examples ./my-examples

Or clone from GitHub:

bash

git clone https://github.com/semanticintent/cal-runtime.git
cd cal-runtime/examples

Examples ​

Overview ​

Running Examples ​

Sample Dataset ​

Entity 1: Tailwind CSS ​

Entity 2: Stack Overflow ​

Entity 3: GitHub Copilot ​

Entity 4: Stable Enterprise Co ​

Example 1: Simple Query ​

Script ​

Concepts Demonstrated ​

Running ​

Expected Output ​

Key Takeaway ​

Example 2: DRIFT Analysis ​

Script ​

Concepts Demonstrated ​

Formula ​

Running ​

Expected Output ​

Key Takeaway ​

Example 3: Tailwind Cascade ​

Script ​

Concepts Demonstrated ​

Dimensions Analyzed ​

Running ​

Expected Results ​

Key Takeaway ​

Example 4: Closed-Loop Pipeline ​

Script ​

The 5-Layer Intelligence Model ​

Layer 1: SENSE ​

Layer 2: ANALYZE ​

Layer 3: MEASURE ​

Layer 4: DECIDE ​

Layer 5: ACT ​

Running ​

Expected Flow ​

Key Takeaway ​

Common Patterns ​

Pattern 1: Query + Filter ​

Pattern 2: Query + DRIFT ​

Pattern 3: Query + FETCH ​

Pattern 4: Query + Monitor ​

Pattern 5: Full Pipeline ​

Modifying Examples ​

Adjust Thresholds ​

Change Dimensions ​

Adjust Cascade Depth ​

Modify DRIFT Values ​

Change FETCH Threshold ​

Creating Custom Examples ​

Step 1: Start Simple ​

Step 2: Add Filtering ​

Step 3: Add Analysis ​

Step 4: Add Measurement ​

Step 5: Add Decision Logic ​

Step 6: Add Output ​

Testing Examples ​

Validate Syntax ​

Dry Run (Analyze) ​

Run with Sample Data ​

Run with Custom Data ​

Next Steps ​

For Learning ​

For Development ​

For Production ​

Additional Resources ​

Example Repository ​

Examples

Overview

Running Examples

Sample Dataset

Entity 1: Tailwind CSS

Entity 2: Stack Overflow

Entity 3: GitHub Copilot

Entity 4: Stable Enterprise Co

Example 1: Simple Query

Script

Concepts Demonstrated

Running

Expected Output

Key Takeaway

Example 2: DRIFT Analysis

Script

Concepts Demonstrated

Formula

Running

Expected Output

Key Takeaway

Example 3: Tailwind Cascade

Script

Concepts Demonstrated

Dimensions Analyzed

Running

Expected Results

Key Takeaway

Example 4: Closed-Loop Pipeline

Script

The 5-Layer Intelligence Model

Layer 1: SENSE

Layer 2: ANALYZE

Layer 3: MEASURE

Layer 4: DECIDE

Layer 5: ACT

Running

Expected Flow

Key Takeaway

Common Patterns

Pattern 1: Query + Filter

Pattern 2: Query + DRIFT

Pattern 3: Query + FETCH

Pattern 4: Query + Monitor

Pattern 5: Full Pipeline

Modifying Examples

Adjust Thresholds

Change Dimensions

Adjust Cascade Depth

Modify DRIFT Values

Change FETCH Threshold

Creating Custom Examples

Step 1: Start Simple

Step 2: Add Filtering

Step 3: Add Analysis

Step 4: Add Measurement

Step 5: Add Decision Logic

Step 6: Add Output

Testing Examples

Validate Syntax

Dry Run (Analyze)

Run with Sample Data

Run with Custom Data

Next Steps

For Learning

For Development

For Production

Additional Resources

Example Repository