Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Plan and verify software releases with versioning, changelogs, release branches, feature flags, canaries, migration gates, rollback, deployment checks, and release readiness. Use when preparing a release, shipping a risky PR, coordinating app/backend/database rollout, recovering from a bad deploy, or defining release policy for a repo.
.claude/skills/majiayu000-release-engineering/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-03 | ✗→✓ | ▲ Improved | -25% | 0% |
| case-05 | ✗→✓ | ▲ Improved | -14% | 0% |
| case-07 | ✗→✓ | ▲ Improved | -22% | 0% |
| case-09 | ✗→✓ | ▲ Improved | -23% | 0% |
| case-10 | ✗→✓ | ▲ Improved | -11% | 0% |
Use this skill to turn "ready to merge" into a controlled release. It connects code, data, config, deployment, monitoring, and rollback.
Classify the release before choosing gates:
| Type | Extra Gates | |---|---| | Patch fix | focused regression test, rollback proof | | Feature release | feature flag, docs/changelog, product acceptance | | API or SDK change | versioning, deprecation, compatibility tests | | Database migration | data-contract-migrations, backup, backfill, recovery | | Config or secret change | config-secrets-environments, rotation, drift check | | Infrastructure change | canary, health checks, capacity and rollback |
For non-trivial releases, require:
Use fresh commands from the current session. Typical gates:
bashgit status --short git log --oneline -n 5 # project-specific build/test command # migration dry run or validation query # deployment health check
Do not claim a release shipped unless remote git, CI, artifact, and runtime state have each been checked at the required level.
textrelease_scope: version_or_identifier: included_changes: excluded_changes: pre_release_checks: rollout_steps: migration_and_config_gates: monitoring: rollback_or_recovery: go_no_go_status:
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-04 | fail→fail | 19,453 | 6,242 | -68% | 1 | 1 | 0% | 1,831 | 1,392 | -24% | 0 | 0 | — |
case-03 | fail→pass | 22,808 | 14,428 | -37% | 1 | 1 | 0% | 3,779 | 2,829 | -25% | 0 | 0 | — |
case-01 | fail→fail | 27,392 | 24,136 | -12% | 1 | 1 | 0% | 4,325 | 3,240 | -25% | 0 | 0 | — |
case-02 | fail→fail | 36,650 | 11,116 | -70% | 1 | 1 | 0% | 5,971 | 2,201 | -63% | 0 | 0 | — |
case-05 | fail→pass | 13,697 | 9,390 | -31% | 1 | 1 | 0% | 2,129 | 1,823 | -14% | 0 | 0 | — |
case-06 | fail→fail | 17,484 | 12,498 | -29% | 1 | 1 | 0% | 2,575 | 2,420 | -6% | 0 | 0 | — |
case-07 | fail→pass | 18,714 | 12,326 | -34% | 1 | 1 | 0% | 2,802 | 2,178 | -22% | 0 | 0 | — |
case-08 | fail→fail | 15,812 | 11,346 | -28% | 1 | 1 | 0% | 2,450 | 2,123 | -13% | 0 | 0 | — |
case-09 | fail→pass | 15,570 | 10,468 | -33% | 1 | 1 | 0% | 2,453 | 1,883 | -23% | 0 | 0 | — |
case-10 | fail→pass | 16,538 | 12,601 | -24% | 1 | 1 | 0% | 2,505 | 2,224 | -11% | 0 | 0 | — |
case-11 | pass→pass | 13,448 | 7,981 | -41% | 1 | 1 | 0% | 2,079 | 1,658 | -20% | 0 | 0 | — |
case-12 | pass→pass | 11,460 | 7,784 | -32% | 1 | 1 | 0% | 1,628 | 1,601 | -2% | 0 | 0 | — |
case-13 | fail→pass | 11,082 | 4,446 | -60% | 1 | 1 | 0% | 1,516 | 1,059 | -30% | 0 | 0 | — |
case-14 | fail→pass | 10,904 | 10,134 | -7% | 1 | 1 | 0% | 1,829 | 2,168 | +19% | 0 | 0 | — |
case-15 | pass→pass | 17,262 | 11,280 | -35% | 1 | 1 | 0% | 2,475 | 2,114 | -15% | 0 | 0 | — |
case-16 | pass→pass | 12,950 | 7,990 | -38% | 1 | 1 | 0% | 2,041 | 1,556 | -24% | 0 | 0 | — |
case-17 | pass→pass | 9,588 | 3,770 | -61% | 1 | 1 | 0% | 1,392 | 991 | -29% | 0 | 0 | — |
case-18 | pass→pass | 14,606 | 9,033 | -38% | 1 | 1 | 0% | 2,045 | 1,804 | -12% | 0 | 0 | — |
case-19 | pass→fail | 9,277 | 3,852 | -58% | 1 | 1 | 0% | 1,354 | 996 | -26% | 0 | 0 | — |
case-20 | pass→pass | 18,114 | 18,456 | +2% | 1 | 1 | 0% | 2,802 | 3,179 | +13% | 0 | 0 | — |
case-21 | pass→pass | 9,332 | 8,487 | -9% | 1 | 1 | 0% | 1,606 | 1,959 | +22% | 0 | 0 | — |
case-22 | pass→pass | 9,709 | 9,058 | -7% | 1 | 1 | 0% | 1,855 | 2,220 | +20% | 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 +27 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.