$npx -y skills add jackspace/ClaudeSkillz --skill checkpoint-workflow-builderBuild resumable state-machine workflows with checkpoint patterns, progress preservation, and automatic recovery for complex multi-phase operations that need to survive interruptions, timeouts, and failures.
| 1 | # Checkpoint Workflow Builder |
| 2 | |
| 3 | Design and implement fault-tolerant workflows that can resume from any point of failure. |
| 4 | |
| 5 | ## Overview |
| 6 | |
| 7 | Complex workflows often fail mid-execution due to: |
| 8 | - Network timeouts |
| 9 | - System crashes |
| 10 | - Resource exhaustion |
| 11 | - External service failures |
| 12 | - User interruptions |
| 13 | |
| 14 | This skill teaches you to build workflows that: |
| 15 | - Save progress at key checkpoints |
| 16 | - Resume from last successful state |
| 17 | - Handle partial failures gracefully |
| 18 | - Provide clear progress visibility |
| 19 | - Enable manual intervention points |
| 20 | |
| 21 | ## When to Use |
| 22 | |
| 23 | Use this skill when: |
| 24 | - Building multi-phase data pipelines |
| 25 | - Implementing long-running migration scripts |
| 26 | - Creating deployment workflows |
| 27 | - Processing large batches of items |
| 28 | - Orchestrating multi-system operations |
| 29 | - Building ETL (Extract, Transform, Load) workflows |
| 30 | - Implementing saga patterns for distributed systems |
| 31 | - Creating user-facing wizards with save/resume |
| 32 | |
| 33 | ## Core Concepts |
| 34 | |
| 35 | ### State Machine Pattern |
| 36 | |
| 37 | ``` |
| 38 | ┌─────────┐ |
| 39 | │ INIT │ |
| 40 | └────┬────┘ |
| 41 | │ |
| 42 | ▼ |
| 43 | ┌────────────┐ |
| 44 | │ DOWNLOAD │ |
| 45 | └─────┬──────┘ |
| 46 | │ |
| 47 | ▼ |
| 48 | ┌────────────┐ |
| 49 | │ PROCESS │ |
| 50 | └─────┬──────┘ |
| 51 | │ |
| 52 | ▼ |
| 53 | ┌────────────┐ |
| 54 | │ VALIDATE │ |
| 55 | └─────┬──────┘ |
| 56 | │ |
| 57 | ▼ |
| 58 | ┌────────────┐ |
| 59 | │ FINALIZE │ |
| 60 | └─────┬──────┘ |
| 61 | │ |
| 62 | ▼ |
| 63 | ┌────────────┐ |
| 64 | │ COMPLETE │ |
| 65 | └────────────┘ |
| 66 | ``` |
| 67 | |
| 68 | Each state: |
| 69 | - Has clear entry conditions |
| 70 | - Performs specific operations |
| 71 | - Saves checkpoint before transition |
| 72 | - Can be resumed independently |
| 73 | |
| 74 | ## Basic Implementation |
| 75 | |
| 76 | ### Simple State Machine |
| 77 | |
| 78 | ```bash |
| 79 | #!/bin/bash |
| 80 | # state-machine.sh - Basic resumable workflow |
| 81 | |
| 82 | STATE_FILE=".workflow_state" |
| 83 | |
| 84 | # Read current state (default: INIT) |
| 85 | CURRENT_STATE=$(cat "$STATE_FILE" 2>/dev/null || echo "INIT") |
| 86 | |
| 87 | echo "Current state: $CURRENT_STATE" |
| 88 | |
| 89 | case "$CURRENT_STATE" in |
| 90 | INIT) |
| 91 | echo "=== Phase 1: Initialization ===" |
| 92 | |
| 93 | # Initialize workspace |
| 94 | mkdir -p workspace |
| 95 | mkdir -p results |
| 96 | |
| 97 | # Download dependencies |
| 98 | echo "Setting up environment..." |
| 99 | |
| 100 | # Save next state |
| 101 | echo "DOWNLOAD" > "$STATE_FILE" |
| 102 | echo "✓ Initialization complete" |
| 103 | echo "Run again to continue" |
| 104 | ;; |
| 105 | |
| 106 | DOWNLOAD) |
| 107 | echo "=== Phase 2: Download Data ===" |
| 108 | |
| 109 | # Download data files |
| 110 | echo "Downloading data..." |
| 111 | # curl -o workspace/data.zip https://example.com/data.zip |
| 112 | |
| 113 | # Verify download |
| 114 | if [ -f "workspace/data.zip" ]; then |
| 115 | echo "EXTRACT" > "$STATE_FILE" |
| 116 | echo "✓ Download complete" |
| 117 | echo "Run again to continue" |
| 118 | else |
| 119 | echo "✗ Download failed - fix and run again" |
| 120 | exit 1 |
| 121 | fi |
| 122 | ;; |
| 123 | |
| 124 | EXTRACT) |
| 125 | echo "=== Phase 3: Extract Data ===" |
| 126 | |
| 127 | # Extract files |
| 128 | echo "Extracting data..." |
| 129 | # unzip workspace/data.zip -d workspace/ |
| 130 | |
| 131 | echo "PROCESS" > "$STATE_FILE" |
| 132 | echo "✓ Extraction complete" |
| 133 | echo "Run again to continue" |
| 134 | ;; |
| 135 | |
| 136 | PROCESS) |
| 137 | echo "=== Phase 4: Process Data ===" |
| 138 | |
| 139 | # Process data |
| 140 | echo "Processing data..." |
| 141 | # ./process_data.sh workspace/ results/ |
| 142 | |
| 143 | echo "VALIDATE" > "$STATE_FILE" |
| 144 | echo "✓ Processing complete" |
| 145 | echo "Run again to continue" |
| 146 | ;; |
| 147 | |
| 148 | VALIDATE) |
| 149 | echo "=== Phase 5: Validate Results ===" |
| 150 | |
| 151 | # Validate results |
| 152 | echo "Validating results..." |
| 153 | # ./validate.sh results/ |
| 154 | |
| 155 | if [ $? -eq 0 ]; then |
| 156 | echo "FINALIZE" > "$STATE_FILE" |
| 157 | echo "✓ Validation passed" |
| 158 | echo "Run again to finalize" |
| 159 | else |
| 160 | echo "✗ Validation failed" |
| 161 | echo "Fix issues and change state to PROCESS to reprocess" |
| 162 | exit 1 |
| 163 | fi |
| 164 | ;; |
| 165 | |
| 166 | FINALIZE) |
| 167 | echo "=== Phase 6: Finalize ===" |
| 168 | |
| 169 | # Cleanup and finalize |
| 170 | echo "Finalizing workflow..." |
| 171 | # mv results/ final/ |
| 172 | # rm -rf workspace/ |
| 173 | |
| 174 | echo "COMPLETE" > "$STATE_FILE" |
| 175 | echo "✓ Workflow complete!" |
| 176 | ;; |
| 177 | |
| 178 | COMPLETE) |
| 179 | echo "=== Workflow Already Complete ===" |
| 180 | echo "Results available in: final/" |
| 181 | ;; |
| 182 | |
| 183 | *) |
| 184 | echo "✗ Unknown state: $CURRENT_STATE" |
| 185 | echo "Reset with: echo 'INIT' > $STATE_FILE" |
| 186 | exit 1 |
| 187 | ;; |
| 188 | esac |
| 189 | ``` |
| 190 | |
| 191 | ### Enhanced with Progress Tracking |
| 192 | |
| 193 | ```bash |
| 194 | #!/bin/bash |
| 195 | # enhanced-state-machine.sh - With detailed progress |
| 196 | |
| 197 | STATE_FILE=".workflow_state" |
| 198 | PROGRESS_FILE=".workflow_progress.json" |
| 199 | |
| 200 | # Initialize progress tracking |
| 201 | init_progress() { |
| 202 | cat > "$PROGRESS_FILE" << EOF |
| 203 | { |
| 204 | "current_state": "INIT", |
| 205 | "started_at": "$(date -Iseconds)", |
| 206 | "updated_at": "$(date -Iseconds)", |
| 207 | "phases": { |
| 208 | "INIT": {"status": "pending", "star |