diff --git a/.github/workflows/deploy.yml b/.github/workflows/deploy.yml
index 244fe26b..7a853dc7 100644
--- a/.github/workflows/deploy.yml
+++ b/.github/workflows/deploy.yml
@@ -76,6 +76,13 @@ jobs:
           role-to-assume: ${{ secrets[format('AWS_ROLE_{0}', matrix.environment)] }}
           aws-region: ${{ env.AWS_REGION }}
 
+      - name: Check SLO Compliance (Production Only)
+        if: matrix.environment == 'prod'
+        run: |
+          node performance/scripts/check-slo-compliance.js
+        env:
+          PROMETHEUS_URL: ${{ secrets.PROMETHEUS_URL }}
+
       - name: Download plan
         uses: actions/download-artifact@v3
         with:
diff --git a/.github/workflows/test.yml b/.github/workflows/test.yml
index 952bc7ec..ac056224 100644
--- a/.github/workflows/test.yml
+++ b/.github/workflows/test.yml
@@ -14,6 +14,27 @@ env:
   COVERAGE_FAIL_LINES_PREDICTIQ_API: "17"
 
 jobs:
+  prometheus-rules-test:
+    name: Prometheus SLO Rules Testing
+    runs-on: ubuntu-latest
+    steps:
+      - uses: actions/checkout@v4
+
+      - name: Install Prometheus tools
+        run: |
+          wget https://github.com/prometheus/prometheus/releases/download/v2.45.0/prometheus-2.45.0.linux-amd64.tar.gz
+          tar xzf prometheus-2.45.0.linux-amd64.tar.gz
+          sudo mv prometheus-2.45.0.linux-amd64/promtool /usr/local/bin/
+
+      - name: Check Prometheus rules syntax
+        run: promtool check rules performance/config/prometheus-slo-rules.yml
+
+      - name: Check alert rules syntax
+        run: promtool check rules performance/config/alerts.yaml
+
+      - name: Run Prometheus rules unit tests
+        run: promtool test rules performance/config/prometheus-slo-rules.test.yml
+
   unit-tests:
     name: Unit Tests
     runs-on: ubuntu-latest
@@ -838,6 +859,7 @@ jobs:
   all-tests-passed:
     name: All Tests Passed
     needs:
+      - prometheus-rules-test
       - unit-tests
       - integration-tests
       - api-rate-limit-tests
diff --git a/contracts/predict-iq/Makefile b/contracts/predict-iq/Makefile
index 76944ad1..42fcd600 100644
--- a/contracts/predict-iq/Makefile
+++ b/contracts/predict-iq/Makefile
@@ -1,6 +1,6 @@
 # PredictIQ Test Suite Makefile
 
-.PHONY: help test test-unit test-integration test-all bench coverage clean install-tools format lint audit
+.PHONY: help test test-unit test-integration test-all bench coverage clean install-tools format lint audit deploy-testnet deploy-mainnet
 
 help:
 	@echo "PredictIQ Test Suite Commands:"
@@ -14,6 +14,8 @@ help:
 	@echo "  make audit             - Run security audit"
 	@echo "  make clean             - Clean build artifacts"
 	@echo "  make install-tools     - Install required tools"
+	@echo "  make deploy-testnet    - Deploy contract to Stellar testnet"
+	@echo "  make deploy-mainnet    - Deploy contract to Stellar mainnet"
 
 # Install required testing tools
 install-tools:
@@ -146,3 +148,47 @@ test-report:
 	@echo "Generating test report..."
 	cargo test --workspace -- --format=json | tee test-report.json
 	@echo "Test report saved to test-report.json"
+
+# Deploy to Stellar testnet
+deploy-testnet:
+	@echo "🚀 Deploying to Stellar Testnet"
+	@echo ""
+	@read -p "Are you sure you want to deploy to TESTNET? [y/N] " confirm; \
+	if [ "$$confirm" = "y" ] || [ "$$confirm" = "Y" ]; then \
+		echo "Proceeding with testnet deployment..."; \
+		cargo build --target wasm32-unknown-unknown --release; \
+		soroban contract deploy \
+			--wasm target/wasm32-unknown-unknown/release/predict_iq.wasm \
+			--source-account $(SOROBAN_ACCOUNT) \
+			--network testnet; \
+		echo "✅ Testnet deployment completed"; \
+	else \
+		echo "❌ Deployment cancelled"; \
+		exit 1; \
+	fi
+
+# Deploy to Stellar mainnet (requires explicit confirmation)
+deploy-mainnet:
+	@echo "🚀 Deploying to Stellar Mainnet"
+	@echo ""
+	@echo "⚠️  WARNING: This will deploy to MAINNET"
+	@echo "⚠️  This action cannot be easily undone"
+	@echo ""
+	@read -p "Are you sure you want to deploy to MAINNET? [y/N] " confirm; \
+	if [ "$$confirm" = "y" ] || [ "$$confirm" = "Y" ]; then \
+		if [ -z "$(CONFIRM_MAINNET)" ] || [ "$(CONFIRM_MAINNET)" != "yes" ]; then \
+			echo "❌ Mainnet deployment requires CONFIRM_MAINNET=yes"; \
+			echo "   Usage: make deploy-mainnet CONFIRM_MAINNET=yes"; \
+			exit 1; \
+		fi; \
+		echo "Proceeding with mainnet deployment..."; \
+		cargo build --target wasm32-unknown-unknown --release; \
+		soroban contract deploy \
+			--wasm target/wasm32-unknown-unknown/release/predict_iq.wasm \
+			--source-account $(SOROBAN_ACCOUNT) \
+			--network mainnet; \
+		echo "✅ Mainnet deployment completed"; \
+	else \
+		echo "❌ Deployment cancelled"; \
+		exit 1; \
+	fi
diff --git a/docs/runbooks/critical-api-response-time.md b/docs/runbooks/critical-api-response-time.md
new file mode 100644
index 00000000..3d098e85
--- /dev/null
+++ b/docs/runbooks/critical-api-response-time.md
@@ -0,0 +1,50 @@
+# Critical API Response Time Runbook
+
+## Alert Meaning
+The API p99 response time has exceeded 500ms critical threshold for 2 minutes.
+
+## Impact
+- Severe user experience degradation
+- Potential request timeouts
+- Risk of cascading failures across dependent services
+
+## Investigation Steps
+
+1. **Immediate assessment**
+   ```bash
+   curl 'http://prometheus:9090/api/v1/query?query=histogram_quantile(0.99,rate(http_request_duration_seconds_bucket[5m]))'
+   ```
+
+2. **Check for errors**
+   ```bash
+   curl 'http://prometheus:9090/api/v1/query?query=rate(http_requests_total{status=~"5.."}[5m])'
+   ```
+
+3. **Identify bottlenecks**
+   - Check database query times
+   - Review external service calls
+   - Check cache hit rates
+
+4. **Check infrastructure**
+   - CPU and memory on all API instances
+   - Network latency
+   - Disk I/O
+
+## Remediation
+
+### Immediate Actions
+1. **Page on-call engineer** - This is a critical issue
+2. **Scale up API instances** if CPU/memory is high
+3. **Check for stuck connections** in database
+4. **Review recent changes** - rollback if necessary
+
+### Emergency Actions
+1. Enable circuit breakers for external services
+2. Reduce cache TTL to force fresh data
+3. Temporarily disable non-critical features
+4. Redirect traffic to backup region if available
+
+### Post-Incident
+1. Conduct root cause analysis
+2. Implement permanent fix
+3. Add performance tests to prevent recurrence
diff --git a/docs/runbooks/critical-database-queries.md b/docs/runbooks/critical-database-queries.md
new file mode 100644
index 00000000..d15d4845
--- /dev/null
+++ b/docs/runbooks/critical-database-queries.md
@@ -0,0 +1,52 @@
+# Critical Database Queries Runbook
+
+## Alert Meaning
+Database p99 query time has exceeded 100ms critical threshold for 2 minutes.
+
+## Impact
+- Severe API latency
+- Risk of connection pool exhaustion
+- Potential cascading failures
+
+## Investigation Steps
+
+1. **Immediate query analysis**
+   ```bash
+   curl 'http://prometheus:9090/api/v1/query?query=histogram_quantile(0.99,rate(db_query_duration_seconds_bucket[5m]))'
+   ```
+
+2. **Check for locks**
+   ```sql
+   SHOW PROCESSLIST;
+   SHOW OPEN TABLES WHERE In_use > 0;
+   ```
+
+3. **Check connection pool**
+   - Active connections
+   - Waiting connections
+   - Connection timeout rate
+
+4. **Review recent changes**
+   - Schema changes
+   - Index changes
+   - Data volume changes
+
+## Remediation
+
+### Immediate Actions
+1. **Page on-call DBA** - This is critical
+2. Kill long-running queries if safe
+3. Check for table locks
+4. Review connection pool settings
+
+### Emergency Actions
+1. Increase connection pool size
+2. Implement query timeout
+3. Redirect traffic if possible
+4. Consider read replicas
+
+### Post-Incident
+1. Conduct root cause analysis
+2. Optimize identified queries
+3. Add query performance tests
+4. Implement query monitoring
diff --git a/docs/runbooks/critical-db-connection-pool-utilization.md b/docs/runbooks/critical-db-connection-pool-utilization.md
new file mode 100644
index 00000000..fee835ba
--- /dev/null
+++ b/docs/runbooks/critical-db-connection-pool-utilization.md
@@ -0,0 +1,47 @@
+# Critical Database Connection Pool Utilization Runbook
+
+## Alert Meaning
+Database connection pool utilization has exceeded 95% critical threshold for 2 minutes.
+
+## Impact
+- Imminent connection pool exhaustion
+- New requests will fail
+- Service outage risk
+
+## Investigation Steps
+
+1. **Immediate assessment**
+   ```bash
+   curl 'http://prometheus:9090/api/v1/query?query=db_connections_active / db_connections_max'
+   ```
+
+2. **Check active connections**
+   ```sql
+   SHOW PROCESSLIST;
+   ```
+
+3. **Identify blocking queries**
+   ```sql
+   SELECT * FROM information_schema.processlist WHERE state != 'Sleep';
+   ```
+
+## Remediation
+
+### Immediate Actions
+1. **Page on-call DBA** - Critical issue
+2. Increase connection pool size immediately
+3. Kill idle connections
+4. Kill long-running queries if safe
+5. Consider read-only mode
+
+### Emergency Actions
+1. Implement connection throttling
+2. Redirect traffic to backup database
+3. Scale database resources
+4. Implement circuit breakers
+
+### Post-Incident
+1. Root cause analysis
+2. Optimize connection usage
+3. Implement connection monitoring
+4. Add connection pool auto-scaling
diff --git a/docs/runbooks/high-api-response-time.md b/docs/runbooks/high-api-response-time.md
new file mode 100644
index 00000000..21e4e2ae
--- /dev/null
+++ b/docs/runbooks/high-api-response-time.md
@@ -0,0 +1,52 @@
+# High API Response Time Runbook
+
+## Alert Meaning
+The API p95 response time has exceeded 200ms threshold for 5 minutes.
+
+## Impact
+- Degraded user experience with slower API responses
+- Potential cascading failures if latency continues to increase
+- May indicate resource contention or inefficient queries
+
+## Investigation Steps
+
+1. **Check current metrics**
+   ```bash
+   curl http://prometheus:9090/api/v1/query?query=histogram_quantile(0.95,rate(http_request_duration_seconds_bucket[5m]))
+   ```
+
+2. **Identify affected endpoints**
+   ```bash
+   curl 'http://prometheus:9090/api/v1/query?query=histogram_quantile(0.95,rate(http_request_duration_seconds_bucket{endpoint!=""}[5m])) by (endpoint)'
+   ```
+
+3. **Check database performance**
+   - Query slow query logs
+   - Check connection pool utilization
+   - Monitor active connections
+
+4. **Check resource utilization**
+   - CPU usage on API servers
+   - Memory usage and GC pauses
+   - Network I/O
+
+5. **Review recent deployments**
+   - Check if any code changes were deployed recently
+   - Review database schema changes
+
+## Remediation
+
+### Immediate Actions
+1. Check if this is a temporary spike or sustained issue
+2. If sustained, consider scaling up API instances
+3. Review and optimize slow queries
+
+### Short-term
+1. Implement query caching if not already in place
+2. Add database indexes for frequently queried fields
+3. Review API endpoint implementations for inefficiencies
+
+### Long-term
+1. Implement performance testing in CI/CD
+2. Set up performance regression alerts
+3. Establish performance SLOs and budgets
diff --git a/docs/runbooks/high-bet-gas-costs.md b/docs/runbooks/high-bet-gas-costs.md
new file mode 100644
index 00000000..84885e99
--- /dev/null
+++ b/docs/runbooks/high-bet-gas-costs.md
@@ -0,0 +1,44 @@
+# High Bet Gas Costs Runbook
+
+## Alert Meaning
+Gas cost for placing bet operations has exceeded 200,000 threshold.
+
+## Impact
+- Increased transaction costs for users
+- Reduced user engagement
+- Competitive disadvantage
+
+## Investigation Steps
+
+1. **Check gas usage**
+   ```bash
+   curl 'http://prometheus:9090/api/v1/query?query=contract_gas_used{operation="place_bet"}'
+   ```
+
+2. **Review contract implementation**
+   - Check for unnecessary operations
+   - Review data validation logic
+   - Check for inefficient storage patterns
+
+3. **Analyze bet patterns**
+   - Check bet complexity
+   - Review market state updates
+
+## Remediation
+
+### Immediate Actions
+1. Analyze recent contract changes
+2. Identify gas optimization opportunities
+3. Review bet validation logic
+
+### Short-term
+1. Optimize contract operations
+2. Reduce validation overhead
+3. Implement efficient data structures
+4. Batch operations where possible
+
+### Long-term
+1. Implement gas monitoring
+2. Add performance benchmarks
+3. Establish gas cost targets
+4. Regular optimization reviews
diff --git a/docs/runbooks/high-contract-gas-costs.md b/docs/runbooks/high-contract-gas-costs.md
new file mode 100644
index 00000000..ed127dff
--- /dev/null
+++ b/docs/runbooks/high-contract-gas-costs.md
@@ -0,0 +1,44 @@
+# High Contract Gas Costs Runbook
+
+## Alert Meaning
+Gas cost for market creation operations has exceeded 500,000 threshold.
+
+## Impact
+- Increased transaction costs
+- Reduced profitability
+- Potential user friction from high fees
+
+## Investigation Steps
+
+1. **Check gas usage trends**
+   ```bash
+   curl 'http://prometheus:9090/api/v1/query?query=contract_gas_used{operation="create_market"}'
+   ```
+
+2. **Review recent contract changes**
+   - Check for new features
+   - Review optimization opportunities
+   - Check for inefficient operations
+
+3. **Analyze transaction patterns**
+   - Check market complexity
+   - Review data storage patterns
+
+## Remediation
+
+### Immediate Actions
+1. Review recent contract deployments
+2. Analyze gas usage by operation
+3. Identify optimization opportunities
+
+### Short-term
+1. Optimize contract code
+2. Reduce storage operations
+3. Batch operations where possible
+4. Implement gas-efficient patterns
+
+### Long-term
+1. Implement gas monitoring
+2. Add gas benchmarks to CI/CD
+3. Establish gas cost SLOs
+4. Regular contract optimization
diff --git a/docs/runbooks/high-cpu-usage.md b/docs/runbooks/high-cpu-usage.md
new file mode 100644
index 00000000..a87849a2
--- /dev/null
+++ b/docs/runbooks/high-cpu-usage.md
@@ -0,0 +1,51 @@
+# High CPU Usage Runbook
+
+## Alert Meaning
+CPU usage has exceeded 80% for 10 minutes.
+
+## Impact
+- Reduced system responsiveness
+- Potential service degradation
+- Risk of timeouts
+
+## Investigation Steps
+
+1. **Check CPU usage**
+   ```bash
+   top -b -n 1 | head -20
+   ```
+
+2. **Identify CPU consumers**
+   ```bash
+   ps aux --sort=-%cpu | head -20
+   ```
+
+3. **Check for runaway processes**
+   - Review application logs
+   - Check for infinite loops
+   - Review recent deployments
+
+4. **Check system load**
+   ```bash
+   uptime
+   ```
+
+## Remediation
+
+### Immediate Actions
+1. Identify CPU-intensive processes
+2. Scale up CPU if possible
+3. Restart services if needed
+4. Check for runaway processes
+
+### Short-term
+1. Optimize CPU usage
+2. Implement CPU limits
+3. Add CPU monitoring
+4. Review code for inefficiencies
+
+### Long-term
+1. Implement performance profiling
+2. Add performance tests to CI/CD
+3. Establish CPU usage SLOs
+4. Regular optimization reviews
diff --git a/docs/runbooks/high-db-connection-pool-utilization.md b/docs/runbooks/high-db-connection-pool-utilization.md
new file mode 100644
index 00000000..5d575d2d
--- /dev/null
+++ b/docs/runbooks/high-db-connection-pool-utilization.md
@@ -0,0 +1,50 @@
+# High Database Connection Pool Utilization Runbook
+
+## Alert Meaning
+Database connection pool utilization has exceeded 80% for 5 minutes.
+
+## Impact
+- Risk of connection exhaustion
+- New requests may fail to acquire connections
+- Potential service degradation
+
+## Investigation Steps
+
+1. **Check connection pool status**
+   ```bash
+   curl 'http://prometheus:9090/api/v1/query?query=db_connections_active / db_connections_max'
+   ```
+
+2. **Identify connection consumers**
+   ```sql
+   SELECT user, COUNT(*) FROM information_schema.processlist GROUP BY user;
+   ```
+
+3. **Check for idle connections**
+   ```sql
+   SELECT * FROM information_schema.processlist WHERE command = 'Sleep';
+   ```
+
+4. **Review query performance**
+   - Check for slow queries holding connections
+   - Look for transaction locks
+
+## Remediation
+
+### Immediate Actions
+1. Increase connection pool size
+2. Kill idle connections
+3. Review and optimize slow queries
+4. Check for connection leaks
+
+### Short-term
+1. Implement connection pooling
+2. Add connection timeout
+3. Optimize query performance
+4. Implement circuit breakers
+
+### Long-term
+1. Monitor connection pool metrics
+2. Establish connection pool SLOs
+3. Implement connection pool auto-scaling
+4. Regular performance tuning
diff --git a/docs/runbooks/high-error-rate.md b/docs/runbooks/high-error-rate.md
new file mode 100644
index 00000000..0e75888e
--- /dev/null
+++ b/docs/runbooks/high-error-rate.md
@@ -0,0 +1,52 @@
+# High Error Rate Runbook
+
+## Alert Meaning
+The API error rate (5xx responses) has exceeded 0.1% for 5 minutes.
+
+## Impact
+- Users experiencing failures
+- Potential data loss or inconsistency
+- Service reliability degradation
+
+## Investigation Steps
+
+1. **Check error rate by endpoint**
+   ```bash
+   curl 'http://prometheus:9090/api/v1/query?query=rate(http_requests_total{status=~"5.."}[5m]) by (endpoint)'
+   ```
+
+2. **Review error logs**
+   ```bash
+   kubectl logs -l app=predictiq-api --tail=1000 | grep ERROR
+   ```
+
+3. **Check specific error types**
+   - 500: Internal Server Error
+   - 502: Bad Gateway
+   - 503: Service Unavailable
+   - 504: Gateway Timeout
+
+4. **Check dependencies**
+   - Database connectivity
+   - Cache availability
+   - External service health
+
+## Remediation
+
+### Immediate Actions
+1. Check application logs for error patterns
+2. Verify database connectivity
+3. Check cache service status
+4. Review recent deployments
+
+### Short-term
+1. Increase logging verbosity if needed
+2. Implement circuit breakers
+3. Add retry logic with exponential backoff
+4. Scale up instances if resource-constrained
+
+### Long-term
+1. Implement comprehensive error tracking
+2. Add error budget monitoring
+3. Establish error rate SLOs
+4. Improve error handling and recovery
diff --git a/docs/runbooks/high-memory-usage.md b/docs/runbooks/high-memory-usage.md
new file mode 100644
index 00000000..b484f987
--- /dev/null
+++ b/docs/runbooks/high-memory-usage.md
@@ -0,0 +1,49 @@
+# High Memory Usage Runbook
+
+## Alert Meaning
+Memory usage has exceeded 90% for 5 minutes.
+
+## Impact
+- Risk of OOM (Out of Memory) errors
+- Service degradation
+- Potential crashes
+
+## Investigation Steps
+
+1. **Check memory usage**
+   ```bash
+   free -h
+   ```
+
+2. **Identify memory consumers**
+   ```bash
+   ps aux --sort=-%mem | head -20
+   ```
+
+3. **Check for memory leaks**
+   - Review application logs
+   - Check for growing memory usage over time
+
+4. **Check cache size**
+   - Redis memory usage
+   - Application cache size
+
+## Remediation
+
+### Immediate Actions
+1. Identify and kill unnecessary processes
+2. Clear caches if safe
+3. Scale up memory if possible
+4. Restart services if needed
+
+### Short-term
+1. Optimize memory usage
+2. Implement memory limits
+3. Add memory monitoring
+4. Review cache policies
+
+### Long-term
+1. Implement memory profiling
+2. Add memory tests to CI/CD
+3. Establish memory SLOs
+4. Regular optimization reviews
diff --git a/docs/runbooks/high-resolution-gas-costs.md b/docs/runbooks/high-resolution-gas-costs.md
new file mode 100644
index 00000000..e0308124
--- /dev/null
+++ b/docs/runbooks/high-resolution-gas-costs.md
@@ -0,0 +1,44 @@
+# High Resolution Gas Costs Runbook
+
+## Alert Meaning
+Gas cost for market resolution operations has exceeded 300,000 threshold.
+
+## Impact
+- Increased operational costs
+- Delayed market resolution
+- Reduced profitability
+
+## Investigation Steps
+
+1. **Check gas usage**
+   ```bash
+   curl 'http://prometheus:9090/api/v1/query?query=contract_gas_used{operation="resolve_market"}'
+   ```
+
+2. **Review resolution logic**
+   - Check for unnecessary state updates
+   - Review settlement calculations
+   - Check for inefficient data access
+
+3. **Analyze market complexity**
+   - Check number of participants
+   - Review outcome determination logic
+
+## Remediation
+
+### Immediate Actions
+1. Analyze resolution operations
+2. Identify optimization opportunities
+3. Review settlement logic
+
+### Short-term
+1. Optimize resolution code
+2. Reduce state updates
+3. Implement efficient calculations
+4. Batch settlement operations
+
+### Long-term
+1. Implement gas monitoring
+2. Add resolution benchmarks
+3. Establish gas cost targets
+4. Regular optimization reviews
diff --git a/docs/runbooks/low-cache-hit-rate.md b/docs/runbooks/low-cache-hit-rate.md
new file mode 100644
index 00000000..c9247caa
--- /dev/null
+++ b/docs/runbooks/low-cache-hit-rate.md
@@ -0,0 +1,45 @@
+# Low Cache Hit Rate Runbook
+
+## Alert Meaning
+Cache hit rate has dropped below 80% for 10 minutes.
+
+## Impact
+- Increased database load
+- Slower response times
+- Higher operational costs
+
+## Investigation Steps
+
+1. **Check cache metrics**
+   ```bash
+   curl 'http://prometheus:9090/api/v1/query?query=rate(cache_hits_total[5m]) / (rate(cache_hits_total[5m]) + rate(cache_misses_total[5m]))'
+   ```
+
+2. **Check cache size and evictions**
+   - Monitor Redis memory usage
+   - Check eviction policy
+   - Review key expiration rates
+
+3. **Identify problematic keys**
+   - Check for cache stampedes
+   - Review key access patterns
+   - Check for inefficient caching
+
+## Remediation
+
+### Immediate Actions
+1. Increase cache size if memory allows
+2. Review cache eviction policy
+3. Check for cache key collisions
+
+### Short-term
+1. Optimize cache key design
+2. Adjust TTL values
+3. Implement cache warming for hot keys
+4. Add cache statistics monitoring
+
+### Long-term
+1. Implement multi-level caching
+2. Use cache-aside pattern more effectively
+3. Add cache performance SLOs
+4. Implement cache coherence strategies
diff --git a/docs/runbooks/low-throughput.md b/docs/runbooks/low-throughput.md
new file mode 100644
index 00000000..7a829282
--- /dev/null
+++ b/docs/runbooks/low-throughput.md
@@ -0,0 +1,51 @@
+# Low Throughput Runbook
+
+## Alert Meaning
+Request throughput has dropped below 1000 req/s for 10 minutes.
+
+## Impact
+- Reduced system capacity
+- Potential service degradation
+- May indicate upstream issues or traffic routing problems
+
+## Investigation Steps
+
+1. **Check current throughput**
+   ```bash
+   curl 'http://prometheus:9090/api/v1/query?query=rate(http_requests_total[5m])'
+   ```
+
+2. **Check by endpoint**
+   ```bash
+   curl 'http://prometheus:9090/api/v1/query?query=rate(http_requests_total[5m]) by (endpoint)'
+   ```
+
+3. **Check for errors**
+   - Are requests being rejected?
+   - Are there connection timeouts?
+   - Check rate limiting status
+
+4. **Check infrastructure**
+   - API instance availability
+   - Load balancer health
+   - Network connectivity
+
+## Remediation
+
+### Immediate Actions
+1. Verify API instances are running
+2. Check load balancer configuration
+3. Verify DNS resolution
+4. Check for rate limiting issues
+
+### Investigation
+1. Review application logs
+2. Check for deployment issues
+3. Verify database connectivity
+4. Check external service dependencies
+
+### Recovery
+1. Restart affected instances if needed
+2. Adjust load balancer configuration
+3. Scale up if capacity is insufficient
+4. Investigate root cause
diff --git a/docs/runbooks/performance-degradation.md b/docs/runbooks/performance-degradation.md
new file mode 100644
index 00000000..e7c1dac9
--- /dev/null
+++ b/docs/runbooks/performance-degradation.md
@@ -0,0 +1,50 @@
+# Performance Degradation Runbook
+
+## Alert Meaning
+Response time has degraded by more than 10% compared to 24 hours ago.
+
+## Impact
+- Degraded user experience
+- Potential SLO violation
+- May indicate resource constraints or code issues
+
+## Investigation Steps
+
+1. **Check current performance**
+   ```bash
+   curl 'http://prometheus:9090/api/v1/query?query=histogram_quantile(0.95,rate(http_request_duration_seconds_bucket[1h]))'
+   ```
+
+2. **Compare with baseline**
+   ```bash
+   curl 'http://prometheus:9090/api/v1/query?query=histogram_quantile(0.95,rate(http_request_duration_seconds_bucket[1h] offset 24h))'
+   ```
+
+3. **Identify affected endpoints**
+   - Check by endpoint
+   - Check by method
+   - Check by status code
+
+4. **Check for recent changes**
+   - Review deployments
+   - Check database changes
+   - Review infrastructure changes
+
+## Remediation
+
+### Immediate Actions
+1. Analyze performance trends
+2. Identify affected endpoints
+3. Check for recent deployments
+
+### Short-term
+1. Optimize identified bottlenecks
+2. Review query performance
+3. Check resource utilization
+4. Consider rollback if recent deployment
+
+### Long-term
+1. Implement performance testing
+2. Add performance regression tests
+3. Establish performance SLOs
+4. Regular optimization reviews
diff --git a/docs/runbooks/service-down.md b/docs/runbooks/service-down.md
new file mode 100644
index 00000000..edeac5f8
--- /dev/null
+++ b/docs/runbooks/service-down.md
@@ -0,0 +1,51 @@
+# Service Down Runbook
+
+## Alert Meaning
+A PredictIQ service has been down for more than 1 minute.
+
+## Impact
+- Service unavailable to users
+- Potential data loss
+- Revenue impact
+
+## Investigation Steps
+
+1. **Check service status**
+   ```bash
+   kubectl get pods -l app=predictiq-api
+   ```
+
+2. **Check service logs**
+   ```bash
+   kubectl logs -l app=predictiq-api --tail=100
+   ```
+
+3. **Check recent events**
+   ```bash
+   kubectl describe pod <pod-name>
+   ```
+
+4. **Check infrastructure**
+   - Node status
+   - Resource availability
+   - Network connectivity
+
+## Remediation
+
+### Immediate Actions
+1. **Page on-call engineer** - Critical issue
+2. Check pod status and restart if needed
+3. Check node health
+4. Review recent deployments
+
+### Emergency Actions
+1. Rollback recent deployment if applicable
+2. Scale up replicas
+3. Check for resource constraints
+4. Failover to backup if available
+
+### Post-Incident
+1. Root cause analysis
+2. Implement monitoring improvements
+3. Add health checks
+4. Improve deployment process
diff --git a/docs/runbooks/significant-performance-degradation.md b/docs/runbooks/significant-performance-degradation.md
new file mode 100644
index 00000000..b4c02253
--- /dev/null
+++ b/docs/runbooks/significant-performance-degradation.md
@@ -0,0 +1,50 @@
+# Significant Performance Degradation Runbook
+
+## Alert Meaning
+Response time has degraded by more than 5% compared to 24 hours ago (alert threshold).
+
+## Impact
+- Noticeable performance degradation
+- Potential SLO impact if trend continues
+- May indicate emerging issues
+
+## Investigation Steps
+
+1. **Check performance trend**
+   ```bash
+   curl 'http://prometheus:9090/api/v1/query_range?query=histogram_quantile(0.95,rate(http_request_duration_seconds_bucket[1h]))&start=<24h-ago>&end=<now>&step=1h'
+   ```
+
+2. **Identify trend direction**
+   - Is degradation continuing?
+   - Is it stabilizing?
+   - Are there spikes?
+
+3. **Check for correlations**
+   - Database performance
+   - Cache hit rates
+   - Resource utilization
+
+4. **Review recent changes**
+   - Code deployments
+   - Infrastructure changes
+   - Data volume changes
+
+## Remediation
+
+### Immediate Actions
+1. Monitor trend closely
+2. Prepare for escalation if continues
+3. Identify potential causes
+
+### Short-term
+1. Optimize identified bottlenecks
+2. Review code changes
+3. Check resource utilization
+4. Consider preventive scaling
+
+### Long-term
+1. Implement continuous performance monitoring
+2. Add performance regression tests
+3. Establish performance budgets
+4. Regular optimization reviews
diff --git a/docs/runbooks/slow-database-queries.md b/docs/runbooks/slow-database-queries.md
new file mode 100644
index 00000000..223e81d7
--- /dev/null
+++ b/docs/runbooks/slow-database-queries.md
@@ -0,0 +1,50 @@
+# Slow Database Queries Runbook
+
+## Alert Meaning
+Database p95 query time has exceeded 50ms for 5 minutes.
+
+## Impact
+- Slower API responses
+- Increased database load
+- Potential connection pool exhaustion
+
+## Investigation Steps
+
+1. **Check slow query log**
+   ```sql
+   SELECT * FROM mysql.slow_log ORDER BY start_time DESC LIMIT 20;
+   ```
+
+2. **Identify slow queries**
+   ```bash
+   curl 'http://prometheus:9090/api/v1/query?query=histogram_quantile(0.95,rate(db_query_duration_seconds_bucket[5m])) by (query)'
+   ```
+
+3. **Check query plans**
+   - Use EXPLAIN for identified slow queries
+   - Check for missing indexes
+   - Review join strategies
+
+4. **Check database health**
+   - Connection pool utilization
+   - Lock contention
+   - Disk I/O
+
+## Remediation
+
+### Immediate Actions
+1. Identify and analyze slow queries
+2. Check for missing indexes
+3. Review query execution plans
+
+### Short-term
+1. Add indexes for frequently queried columns
+2. Optimize query logic
+3. Implement query result caching
+4. Consider query rewriting
+
+### Long-term
+1. Implement query performance monitoring
+2. Add database performance tests
+3. Establish query performance SLOs
+4. Regular index maintenance
diff --git a/performance/config/alerts.yaml b/performance/config/alerts.yaml
index 093bc7a4..778ff282 100644
--- a/performance/config/alerts.yaml
+++ b/performance/config/alerts.yaml
@@ -14,6 +14,7 @@ groups:
         annotations:
           summary: "High API response time detected"
           description: "API p95 response time is {{ $value }}s, exceeding 200ms threshold"
+          runbook_url: "https://docs.predictiq.com/runbooks/high-api-response-time"
 
       - alert: CriticalAPIResponseTime
         expr: histogram_quantile(0.99, rate(http_request_duration_seconds_bucket[5m])) > 0.5
@@ -24,6 +25,7 @@ groups:
         annotations:
           summary: "Critical API response time detected"
           description: "API p99 response time is {{ $value }}s, exceeding 500ms critical threshold"
+          runbook_url: "https://docs.predictiq.com/runbooks/critical-api-response-time"
 
       - alert: HighErrorRate
         expr: rate(http_requests_total{status=~"5.."}[5m]) / rate(http_requests_total[5m]) > 0.001
@@ -34,6 +36,7 @@ groups:
         annotations:
           summary: "High error rate detected"
           description: "Error rate is {{ $value | humanizePercentage }}, exceeding 0.1% threshold"
+          runbook_url: "https://docs.predictiq.com/runbooks/high-error-rate"
 
       - alert: LowThroughput
         expr: rate(http_requests_total[5m]) < 1000
@@ -44,6 +47,7 @@ groups:
         annotations:
           summary: "Low request throughput"
           description: "Request throughput is {{ $value }} req/s, below 1000 req/s minimum"
+          runbook_url: "https://docs.predictiq.com/runbooks/low-throughput"
 
   - name: cache_performance
     interval: 30s
@@ -57,6 +61,7 @@ groups:
         annotations:
           summary: "Low cache hit rate"
           description: "Cache hit rate is {{ $value | humanizePercentage }}, below 80% threshold"
+          runbook_url: "https://docs.predictiq.com/runbooks/low-cache-hit-rate"
 
   - name: database_performance
     interval: 30s
@@ -70,6 +75,7 @@ groups:
         annotations:
           summary: "Slow database queries detected"
           description: "Database p95 query time is {{ $value }}s, exceeding 50ms threshold"
+          runbook_url: "https://docs.predictiq.com/runbooks/slow-database-queries"
 
       - alert: CriticalDatabaseQueries
         expr: histogram_quantile(0.99, rate(db_query_duration_seconds_bucket[5m])) > 0.1
@@ -80,6 +86,7 @@ groups:
         annotations:
           summary: "Critical database query performance"
           description: "Database p99 query time is {{ $value }}s, exceeding 100ms critical threshold"
+          runbook_url: "https://docs.predictiq.com/runbooks/critical-database-queries"
 
       - alert: HighDBConnectionPoolUtilization
         expr: db_connections_active / db_connections_max > 0.8
@@ -90,6 +97,7 @@ groups:
         annotations:
           summary: "High database connection pool utilization"
           description: "Connection pool utilization is {{ $value | humanizePercentage }}, exceeding 80% threshold"
+          runbook_url: "https://docs.predictiq.com/runbooks/high-db-connection-pool-utilization"
 
       - alert: CriticalDBConnectionPoolUtilization
         expr: db_connections_active / db_connections_max > 0.95
@@ -100,6 +108,7 @@ groups:
         annotations:
           summary: "Critical database connection pool utilization"
           description: "Connection pool utilization is {{ $value | humanizePercentage }}, exceeding 95% critical threshold"
+          runbook_url: "https://docs.predictiq.com/runbooks/critical-db-connection-pool-utilization"
 
   - name: contract_performance
     interval: 1m
@@ -113,6 +122,7 @@ groups:
         annotations:
           summary: "High gas costs for market creation"
           description: "Create market gas cost is {{ $value }}, exceeding 500000 threshold"
+          runbook_url: "https://docs.predictiq.com/runbooks/high-contract-gas-costs"
 
       - alert: HighBetGasCosts
         expr: contract_gas_used{operation="place_bet"} > 200000
@@ -123,6 +133,7 @@ groups:
         annotations:
           summary: "High gas costs for placing bets"
           description: "Place bet gas cost is {{ $value }}, exceeding 200000 threshold"
+          runbook_url: "https://docs.predictiq.com/runbooks/high-bet-gas-costs"
 
       - alert: HighResolutionGasCosts
         expr: contract_gas_used{operation="resolve_market"} > 300000
@@ -133,6 +144,7 @@ groups:
         annotations:
           summary: "High gas costs for market resolution"
           description: "Resolve market gas cost is {{ $value }}, exceeding 300000 threshold"
+          runbook_url: "https://docs.predictiq.com/runbooks/high-resolution-gas-costs"
 
   - name: system_health
     interval: 30s
@@ -146,6 +158,7 @@ groups:
         annotations:
           summary: "Service is down"
           description: "{{ $labels.job }} has been down for more than 1 minute"
+          runbook_url: "https://docs.predictiq.com/runbooks/service-down"
 
       - alert: HighMemoryUsage
         expr: (node_memory_MemTotal_bytes - node_memory_MemAvailable_bytes) / node_memory_MemTotal_bytes > 0.9
@@ -156,6 +169,7 @@ groups:
         annotations:
           summary: "High memory usage"
           description: "Memory usage is {{ $value | humanizePercentage }}, exceeding 90% threshold"
+          runbook_url: "https://docs.predictiq.com/runbooks/high-memory-usage"
 
       - alert: HighCPUUsage
         expr: 100 - (avg by (instance) (rate(node_cpu_seconds_total{mode="idle"}[5m])) * 100) > 80
@@ -166,6 +180,7 @@ groups:
         annotations:
           summary: "High CPU usage"
           description: "CPU usage is {{ $value }}%, exceeding 80% threshold"
+          runbook_url: "https://docs.predictiq.com/runbooks/high-cpu-usage"
 
   - name: tts_quota
     interval: 1m
@@ -227,6 +242,7 @@ groups:
         annotations:
           summary: "Performance degradation detected"
           description: "Response time has degraded by more than 10% compared to 24h ago"
+          runbook_url: "https://docs.predictiq.com/runbooks/performance-degradation"
 
       - alert: SignificantPerformanceDegradation
         expr: |
@@ -242,3 +258,4 @@ groups:
         annotations:
           summary: "Performance degradation alert threshold reached"
           description: "Response time has degraded by more than 5% compared to 24h ago (alert threshold)"
+          runbook_url: "https://docs.predictiq.com/runbooks/significant-performance-degradation"
diff --git a/performance/config/prometheus-slo-rules.test.yml b/performance/config/prometheus-slo-rules.test.yml
new file mode 100644
index 00000000..17bf5cf7
--- /dev/null
+++ b/performance/config/prometheus-slo-rules.test.yml
@@ -0,0 +1,112 @@
+# Unit tests for Prometheus SLO rules
+tests:
+  - interval: 1m
+    input_series:
+      - series: 'http_requests_total{status="200"}'
+        values: '100+100x10'
+      - series: 'http_requests_total{status="500"}'
+        values: '1+0x10'
+      - series: 'http_request_duration_seconds_bucket{le="0.2"}'
+        values: '95+5x10'
+      - series: 'http_request_duration_seconds_bucket{le="+Inf"}'
+        values: '100+10x10'
+      - series: 'db_query_duration_seconds_bucket{le="0.05"}'
+        values: '90+5x10'
+      - series: 'db_query_duration_seconds_bucket{le="+Inf"}'
+        values: '100+10x10'
+      - series: 'redis_commands_total{status="success"}'
+        values: '1000+100x10'
+      - series: 'redis_commands_total{status="failed"}'
+        values: '1+0x10'
+
+  - interval: 5m
+    input_series:
+      - series: 'http_requests_total{status="200"}'
+        values: '500+100x10'
+      - series: 'http_requests_total{status="500"}'
+        values: '5+1x10'
+    promql_expr_test:
+      - expr: 'sum(rate(http_requests_total{status=~"2..|3.."}[5m])) / sum(rate(http_requests_total[5m])) * 100'
+        expected_samples:
+          - labels: '{}'
+            value: '99'
+
+  - interval: 1h
+    input_series:
+      - series: 'http_requests_total{status="200"}'
+        values: '3600+100x24'
+      - series: 'http_requests_total{status="500"}'
+        values: '36+1x24'
+    promql_expr_test:
+      - expr: 'slo:api_availability:success_rate'
+        expected_samples:
+          - labels: '{}'
+            value: '99'
+
+  - interval: 1m
+    input_series:
+      - series: 'http_requests_total{status="200"}'
+        values: '100+0x10'
+      - series: 'http_requests_total{status="500"}'
+        values: '100+0x10'
+    promql_expr_test:
+      - expr: 'sum(rate(http_requests_total{status=~"2..|3.."}[5m])) / sum(rate(http_requests_total[5m])) * 100'
+        expected_samples:
+          - labels: '{}'
+            value: '50'
+
+  - interval: 1m
+    input_series:
+      - series: 'http_request_duration_seconds_bucket{le="0.2"}'
+        values: '50+0x10'
+      - series: 'http_request_duration_seconds_bucket{le="+Inf"}'
+        values: '100+0x10'
+    promql_expr_test:
+      - expr: 'histogram_quantile(0.95, sum(rate(http_request_duration_seconds_bucket[5m])) by (le)) <= 0.2'
+        expected_samples:
+          - labels: '{}'
+            value: '1'
+
+  - interval: 1m
+    input_series:
+      - series: 'redis_commands_total{status="success"}'
+        values: '1000+0x10'
+      - series: 'redis_commands_total{status="failed"}'
+        values: '0+0x10'
+    promql_expr_test:
+      - expr: 'sum(rate(redis_commands_total{status="success"}[5m])) / sum(rate(redis_commands_total[5m])) * 100'
+        expected_samples:
+          - labels: '{}'
+            value: '100'
+
+  - interval: 1m
+    input_series:
+      - series: 'http_requests_total{status="200"}'
+        values: '100+0x10'
+      - series: 'http_requests_total{status="500"}'
+        values: '1+0x10'
+    alert_rule_test:
+      - alert_rule_name: 'SLOFastBurn'
+        eval_time: '10m'
+        expected_alerts:
+          - exp_labels:
+              severity: 'critical'
+              slo: 'api_availability'
+            exp_annotations:
+              summary: 'SLO fast burn detected for API availability'
+
+  - interval: 1m
+    input_series:
+      - series: 'http_requests_total{status="200"}'
+        values: '100+0x10'
+      - series: 'http_requests_total{status="500"}'
+        values: '100+0x10'
+    alert_rule_test:
+      - alert_rule_name: 'SLOErrorBudgetExhausted'
+        eval_time: '10m'
+        expected_alerts:
+          - exp_labels:
+              severity: 'critical'
+              slo: 'api_availability'
+            exp_annotations:
+              summary: 'SLO error budget exhausted for API availability'
diff --git a/performance/scripts/check-slo-compliance.js b/performance/scripts/check-slo-compliance.js
new file mode 100755
index 00000000..849780a1
--- /dev/null
+++ b/performance/scripts/check-slo-compliance.js
@@ -0,0 +1,112 @@
+#!/usr/bin/env node
+
+/**
+ * Check SLO compliance before deployment
+ * Queries Prometheus for current SLO metrics and blocks deployment if error budget exhausted
+ */
+
+const https = require('https');
+const fs = require('fs');
+const path = require('path');
+
+const PROMETHEUS_URL = process.env.PROMETHEUS_URL || 'http://prometheus:9090';
+const SLO_CONFIG = path.join(__dirname, '../config/slo.json');
+const ALLOW_OVERRIDE = process.env.CONFIRM_SLO_OVERRIDE === 'yes';
+
+async function queryPrometheus(query) {
+  return new Promise((resolve, reject) => {
+    const url = new URL(`${PROMETHEUS_URL}/api/v1/query`);
+    url.searchParams.append('query', query);
+
+    const protocol = url.protocol === 'https:' ? https : require('http');
+    
+    protocol.get(url, (res) => {
+      let data = '';
+      res.on('data', chunk => data += chunk);
+      res.on('end', () => {
+        try {
+          const result = JSON.parse(data);
+          if (result.status === 'success') {
+            resolve(result.data.result);
+          } else {
+            reject(new Error(`Prometheus error: ${result.error}`));
+          }
+        } catch (e) {
+          reject(e);
+        }
+      });
+    }).on('error', reject);
+  });
+}
+
+async function checkSLOCompliance() {
+  try {
+    const sloConfig = JSON.parse(fs.readFileSync(SLO_CONFIG, 'utf8'));
+    
+    console.log('🔍 Checking SLO compliance...\n');
+    
+    let allCompliant = true;
+    const results = [];
+
+    for (const [sloName, sloTarget] of Object.entries(sloConfig.slos)) {
+      try {
+        // Query error budget remaining
+        const query = `slo:${sloName}:error_budget_remaining`;
+        const result = await queryPrometheus(query);
+        
+        if (result.length === 0) {
+          console.warn(`⚠️  No data for SLO: ${sloName}`);
+          continue;
+        }
+
+        const errorBudgetRemaining = parseFloat(result[0].value[1]);
+        const compliant = errorBudgetRemaining > 0;
+        
+        results.push({
+          slo: sloName,
+          target: sloTarget,
+          errorBudgetRemaining,
+          compliant
+        });
+
+        const icon = compliant ? '✅' : '❌';
+        console.log(`${icon} ${sloName}`);
+        console.log(`   Target: ${sloTarget}%`);
+        console.log(`   Error Budget Remaining: ${errorBudgetRemaining.toFixed(2)}%\n`);
+
+        if (!compliant) {
+          allCompliant = false;
+        }
+      } catch (error) {
+        console.error(`❌ Error checking ${sloName}: ${error.message}`);
+        allCompliant = false;
+      }
+    }
+
+    // Print summary
+    console.log('─'.repeat(50));
+    if (allCompliant) {
+      console.log('✅ All SLOs compliant - deployment allowed\n');
+      return true;
+    } else {
+      console.log('❌ SLO compliance check failed\n');
+      
+      if (ALLOW_OVERRIDE) {
+        console.log('⚠️  Override enabled via CONFIRM_SLO_OVERRIDE=yes');
+        console.log('   Proceeding with deployment (requires approval)\n');
+        return true;
+      } else {
+        console.log('🛑 Deployment blocked - error budget exhausted');
+        console.log('   Set CONFIRM_SLO_OVERRIDE=yes to override (requires approval)\n');
+        return false;
+      }
+    }
+  } catch (error) {
+    console.error(`Fatal error: ${error.message}`);
+    process.exit(1);
+  }
+}
+
+checkSLOCompliance().then(success => {
+  process.exit(success ? 0 : 1);
+});