Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Get the result of an Output SDK workflow execution. Use when retrieving the output of a completed workflow, getting the return value, or checking what a workflow produced after async execution.
.claude/skills/growthxai-output-workflow-result/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-02 | ✗→✓ | ▲ Improved | 337% | 0% |
| case-04 | ✗→✓ | ▲ Improved | 31% | 0% |
| case-05 | ✗→✓ | ▲ Improved | 24% | 0% |
| case-06 | ✗→✓ | ▲ Improved | 15% | 0% |
| case-07 | ✗→✓ | ▲ Improved | 5% | 0% |
This skill retrieves the result (return value) of a completed workflow execution. Use it after a workflow started with npx output workflow start has completed, or to retrieve results from historical runs.
npx output workflow debug insteadbashnpx output workflow result <workflowId>
The result is the return value of the workflow function, typically JSON.
Before getting results, verify the workflow completed:
bashnpx output workflow status <workflowId> # Should show: COMPLETED npx output workflow result <workflowId>
A successful workflow returns the value from its fn function:
typescript// Workflow code export default workflow( { fn: async input => { return { processed: true, count: 42 }; } } );
bash# Result output npx output workflow result abc123 # { "processed": true, "count": 42 }
If you try to get the result of a failed workflow:
npx output workflow debug instead to see what went wrongSome workflows don't return a value:
bashnpx output workflow result abc123 # null
Scenario: Get result after async start
bash# Start workflow npx output workflow start calculate --input '{"values": [1, 2, 3]}' # Output: Workflow ID: calc-abc123 # Wait for completion npx output workflow status calc-abc123 # Status: COMPLETED # Get the result npx output workflow result calc-abc123 # { "sum": 6, "average": 2 }
Scenario: Process result with jq
bash# Extract specific field npx output workflow result abc123 | jq '.total' # Format for display npx output workflow result abc123 | jq '.' # Save to file npx output workflow result abc123 > result.json
Scenario: Compare results between runs
bash# Get results from two runs npx output workflow result run-1-abc > result1.json npx output workflow result run-2-xyz > result2.json # Compare diff result1.json result2.json
Scenario: Use in a script
bashWORKFLOW_ID="abc123" # Wait for completion while [[ $(npx output workflow status $WORKFLOW_ID) == *"RUNNING"* ]]; do sleep 5 done # Check if completed successfully if [[ $(npx output workflow status $WORKFLOW_ID) == *"COMPLETED"* ]]; then RESULT=$(npx output workflow result $WORKFLOW_ID) echo "Workflow result: $RESULT" else echo "Workflow did not complete successfully" npx output workflow debug $WORKFLOW_ID fi
bashnpx output workflow result abc123 # { "key": "value", "nested": { "data": true } }
bashnpx output workflow result abc123 # [1, 2, 3, 4, 5]
bashnpx output workflow result abc123 # 42 npx output workflow result abc123 # "success" npx output workflow result abc123 # true
For large results, redirect to a file:
bashnpx output workflow result abc123 > large-result.json
npx output workflow runs list to find valid IDsnpx output workflow status <id>npx output workflow debugnpx output workflow status <id> - Check if workflow completednpx output workflow debug <id> - Debug failed workflowsnpx output workflow run <name> - Run and get result in one stepnpx output workflow runs list - Find workflow IDs| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→fail | 5,418 | 1,761 | -67% | 1 | 1 | 0% | 251 | 1,372 | +447% | 0 | 0 | — |
case-02 | fail→pass | 2,773 | 4,390 | +58% | 1 | 1 | 0% | 417 | 1,822 | +337% | 0 | 0 | — |
case-03 | fail→fail | 5,646 | 1,490 | -74% | 1 | 1 | 0% | 802 | 1,309 | +63% | 0 | 0 | — |
case-04 | fail→pass | 8,195 | 3,502 | -57% | 1 | 1 | 0% | 1,313 | 1,719 | +31% | 0 | 0 | — |
case-05 | fail→pass | 7,782 | 2,663 | -66% | 1 | 1 | 0% | 1,270 | 1,572 | +24% | 0 | 0 | — |
case-06 | fail→pass | 7,290 | 2,061 | -72% | 1 | 1 | 0% | 1,275 | 1,469 | +15% | 0 | 0 | — |
case-07 | fail→pass | 8,985 | 2,106 | -77% | 1 | 1 | 0% | 1,366 | 1,432 | +5% | 0 | 0 | — |
case-08 | fail→pass | 6,733 | 4,106 | -39% | 1 | 1 | 0% | 1,089 | 1,790 | +64% | 0 | 0 | — |
case-09 | fail→pass | 8,318 | 5,198 | -38% | 1 | 1 | 0% | 1,452 | 2,049 | +41% | 0 | 0 | — |
case-10 | pass→pass | 11,981 | 7,604 | -37% | 1 | 1 | 0% | 1,711 | 2,432 | +42% | 0 | 0 | — |
case-11 | fail→pass | 9,662 | 1,943 | -80% | 1 | 1 | 0% | 1,438 | 1,364 | -5% | 0 | 0 | — |
case-12 | fail→pass | 10,269 | 4,403 | -57% | 1 | 1 | 0% | 1,624 | 1,838 | +13% | 0 | 0 | — |
case-13 | pass→pass | 8,238 | 3,520 | -57% | 1 | 1 | 0% | 1,341 | 1,728 | +29% | 0 | 0 | — |
case-14 | fail→pass | 8,983 | 2,118 | -76% | 1 | 1 | 0% | 1,393 | 1,448 | +4% | 0 | 0 | — |
case-15 | fail→pass | 10,303 | 2,850 | -72% | 1 | 1 | 0% | 1,624 | 1,590 | -2% | 0 | 0 | — |
case-16 | pass→pass | 10,996 | 3,123 | -72% | 1 | 1 | 0% | 1,565 | 1,647 | +5% | 0 | 0 | — |
case-17 | fail→pass | 8,242 | 1,551 | -81% | 1 | 1 | 0% | 1,338 | 1,311 | -2% | 0 | 0 | — |
case-18 | pass→pass | 7,492 | 3,003 | -60% | 1 | 1 | 0% | 1,241 | 1,622 | +31% | 0 | 0 | — |
case-19 | fail→pass | 8,140 | 3,076 | -62% | 1 | 1 | 0% | 1,353 | 1,577 | +17% | 0 | 0 | — |
case-20 | fail→pass | 8,459 | 3,209 | -62% | 1 | 1 | 0% | 1,442 | 1,623 | +13% | 0 | 0 | — |
case-21 | fail→pass | 9,073 | 2,243 | -75% | 1 | 1 | 0% | 1,650 | 1,510 | -8% | 0 | 0 | — |
case-22 | fail→pass | 6,672 | 1,516 | -77% | 1 | 1 | 0% | 1,007 | 1,356 | +35% | 0 | 0 | — |
case-23 | fail→pass | 5,801 | 2,553 | -56% | 1 | 1 | 0% | 938 | 1,535 | +64% | 0 | 0 | — |
DecimalAI ran this skill against gemini-3.6-flash twice over the same eval suite — once with the skill loaded and once without — and compared the two runs case by case. 23 cases were attempted, and 22 counted toward the lift figure. The other 1 produced results that are not comparable between the two arms, so they are excluded from the headline rather than averaged into it. The headline lift of +74 percentage points is the difference between those two pass rates over the 22 comparable cases.
Without the skill loaded, the model failed this case. With it loaded, the same prompt on the same model passed. This is one improved case from the latest verified run; every case, including any that regressed, is in the table above.
Other measured skills in the registry, with their headline benchmark lift.