Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Deployment workflows, pre-deploy validation, smoke testing, and rollback procedures. Use when deploying to staging or production, running smoke tests, validating deployments, or planning rollback. Do NOT use for local development or CI pipeline configuration.
.claude/skills/bybren-llc-deployment-sop/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-03 | ✗→✓ | ▲ Improved | 14% | 0% |
| case-13 | ✗→✓ | ▲ Improved | -26% | 0% |
| case-02 | ✗→✓ | ▲ Improved | 13% | 0% |
| case-07 | ✗→✓ | ▲ Improved | -23% | 0% |
| case-08 | ✗→✓ | ▲ Improved | 82% | 0% |
Route to existing deployment SOPs and provide checklists for safe, validated deployments. This skill does NOT duplicate SOP content—it links to authoritative sources.
| Document | Location | Purpose | | ------------------------ | ------------------------------------------------- | --------------------------- | | Semantic Release SOP | docs/ci-cd/Semantic-Release-Deployment-SOP.md | Release automation workflow | | Staging/UAT Release SOP | docs/sop/STAGING-UAT-RELEASE-SOP.md | UAT validation process | | Linux Dev Machine Access | docs/deployment/LINUX-DEV-MACHINE-ACCESS-SOP.md | Dev server access | | Production Server Access | docs/deployment/PRODUCTION-SERVER-ACCESS-SOP.md | Production deployment |
Before ANY deployment:
bash# Validate before deploy yarn ci:validate yarn build
After deployment completes:
curl https://{domain}/api/healthbash# Smoke test commands curl -s https://{domain}/api/health | jq . # Expected: {"status":"healthy","timestamp":"..."}
For Linear ticket attachment:
markdown## Deployment Evidence - {{TICKET_PREFIX}}-XXX ### Environment - **Target**: Staging / Production - **Branch**: `{branch_name}` - **Commit**: `{commit_sha}` ### Pre-Deployment - [x] CI checks passed - [x] PR merged - [x] Migrations verified ### Post-Deployment - [x] Health check: PASSED - [x] Auth flow: PASSED - [x] Smoke tests: PASSED ### Verification curl -s https://{domain}/api/health {"status":"healthy","timestamp":"2025-XX-XXTXX:XX:XX.XXXZ"}
If deployment fails:
git revert {commit_sha}| Branch | Environment | Auto-Deploy | | -------- | ----------- | ----------- | | dev | Staging | Manual pull | | main | Production | Automatic |
Note: Merging to dev builds Docker image but requires manual pull on staging server.
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-03 | fail→pass | 10,242 | 5,206 | -49% | 1 | 1 | 0% | 1,583 | 1,808 | +14% | 0 | 0 | — |
case-13 | fail→pass | 14,221 | 4,644 | -67% | 1 | 1 | 0% | 2,054 | 1,521 | -26% | 0 | 0 | — |
case-01 | pass→pass | 9,596 | 4,661 | -51% | 1 | 1 | 0% | 1,718 | 1,621 | -6% | 0 | 0 | — |
case-02 | fail→pass | 11,977 | 7,104 | -41% | 1 | 1 | 0% | 1,767 | 2,003 | +13% | 0 | 0 | — |
case-04 | pass→pass | 5,287 | 4,904 | -7% | 1 | 1 | 0% | 997 | 1,727 | +73% | 0 | 0 | — |
case-05 | pass→pass | 14,926 | 12,057 | -19% | 1 | 1 | 0% | 2,427 | 2,913 | +20% | 0 | 0 | — |
case-06 | pass→fail | 6,666 | 4,817 | -28% | 1 | 1 | 0% | 1,305 | 1,744 | +34% | 0 | 0 | — |
case-07 | fail→pass | 8,860 | 2,073 | -77% | 1 | 1 | 0% | 1,400 | 1,079 | -23% | 0 | 0 | — |
case-08 | fail→pass | 4,225 | 1,802 | -57% | 1 | 1 | 0% | 619 | 1,128 | +82% | 0 | 0 | — |
case-09 | fail→pass | 9,521 | 3,071 | -68% | 1 | 1 | 0% | 793 | 1,014 | +28% | 0 | 0 | — |
case-10 | fail→pass | 15,697 | 2,972 | -81% | 1 | 1 | 0% | 2,702 | 1,330 | -51% | 0 | 0 | — |
case-11 | pass→pass | 14,481 | 3,465 | -76% | 1 | 1 | 0% | 2,153 | 1,338 | -38% | 0 | 0 | — |
case-12 | fail→pass | 10,050 | 2,807 | -72% | 1 | 1 | 0% | 1,868 | 1,252 | -33% | 0 | 0 | — |
case-14 | fail→pass | 2,928 | 2,074 | -29% | 1 | 1 | 0% | 433 | 1,083 | +150% | 0 | 0 | — |
case-15 | fail→pass | 17,374 | 6,527 | -62% | 1 | 1 | 0% | 2,643 | 1,931 | -27% | 0 | 0 | — |
case-16 | fail→pass | 10,947 | 10,542 | -4% | 1 | 1 | 0% | 1,562 | 1,394 | -11% | 0 | 0 | — |
case-17 | fail→pass | 13,087 | 4,051 | -69% | 1 | 1 | 0% | 1,817 | 1,459 | -20% | 0 | 0 | — |
case-18 | fail→pass | 12,450 | 2,749 | -78% | 1 | 1 | 0% | 1,925 | 1,284 | -33% | 0 | 0 | — |
case-19 | pass→pass | 10,612 | 4,270 | -60% | 1 | 1 | 0% | 1,500 | 1,439 | -4% | 0 | 0 | — |
case-20 | pass→pass | 6,282 | 2,471 | -61% | 1 | 1 | 0% | 893 | 1,128 | +26% | 0 | 0 | — |
case-21 | fail→pass | 18,172 | 3,136 | -83% | 1 | 1 | 0% | 2,383 | 1,236 | -48% | 0 | 0 | — |
case-22 | pass→pass | 13,505 | 4,721 | -65% | 1 | 1 | 0% | 1,984 | 1,523 | -23% | 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. 22 cases were attempted. The headline lift of +59 percentage points is the difference between those two pass rates over the 22 comparable cases. 1 case got worse with the skill loaded, and it is included in that figure.
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.