Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Search and deploy services from Railway's template marketplace. Use when user wants to add a service from a template, find templates for a specific use case, or deploy tools like Ghost, Strapi, n8n, Minio, Uptime Kuma, etc. For databases (Postgres, Redis, MySQL, MongoDB), prefer the railway-database skill.
.claude/skills/davila7-railway-templates/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-05 | ✗→✓ | ▲ Improved | 16% | 0% |
| case-07 | ✗→✓ | ▲ Improved | -42% | 0% |
| case-22 | ✗→✓ | ▲ Improved | 18% | 0% |
| case-12 | ✗→✓ | ▲ Improved | -15% | 0% |
| case-15 | ✗→✓ | ▲ Improved | -7% | 0% |
Search and deploy services from Railway's template marketplace.
| Category | Template | Code | |----------|----------|------| | Databases | PostgreSQL | postgres | | | Redis | redis | | | MySQL | mysql | | | MongoDB | mongodb | | CMS | Ghost | ghost | | | Strapi | strapi | | Storage | Minio | minio | | Automation | n8n | n8n | | Monitoring | Uptime Kuma | uptime-kuma |
For other templates, use the search query below.
Get project context:
bashrailway status --json
Extract:
id - project IDenvironments.edges[0].node.id - environment IDGet workspace ID:
bashbash <<'SCRIPT' ${CLAUDE_PLUGIN_ROOT}/skills/lib/railway-api.sh \ 'query getWorkspace($projectId: String!) { project(id: $projectId) { workspaceId } }' \ '{"projectId": "PROJECT_ID"}' SCRIPT
List available templates with optional filters:
bashbash <<'SCRIPT' ${CLAUDE_PLUGIN_ROOT}/skills/lib/railway-api.sh \ 'query templates($first: Int, $verified: Boolean) { templates(first: $first, verified: $verified) { edges { node { name code description category } } } }' \ '{"first": 20, "verified": true}' SCRIPT
| Argument | Type | Description | |----------|------|-------------| | first | Int | Number of results (max ~100) | | verified | Boolean | Only verified templates | | recommended | Boolean | Only recommended templates |
10 requests per minute. Don't spam searches.
Fetch a specific template by code:
bashbash <<'SCRIPT' ${CLAUDE_PLUGIN_ROOT}/skills/lib/railway-api.sh \ 'query template($code: String!) { template(code: $code) { id name description serializedConfig } }' \ '{"code": "postgres"}' SCRIPT
Returns:
id - template ID (needed for deployment)serializedConfig - service configuration (needed for deployment)bashbash <<'SCRIPT' ${CLAUDE_PLUGIN_ROOT}/skills/lib/railway-api.sh \ 'query template($code: String!) { template(code: $code) { id serializedConfig } }' \ '{"code": "postgres"}' SCRIPT
bashbash <<'SCRIPT' ${CLAUDE_PLUGIN_ROOT}/skills/lib/railway-api.sh \ 'mutation deployTemplate($input: TemplateDeployV2Input!) { templateDeployV2(input: $input) { projectId workflowId } }' \ '{ "input": { "templateId": "TEMPLATE_ID_FROM_STEP_1", "serializedConfig": SERIALIZED_CONFIG_FROM_STEP_1, "projectId": "PROJECT_ID", "environmentId": "ENVIRONMENT_ID", "workspaceId": "WORKSPACE_ID" } }' SCRIPT
Important: serializedConfig is the exact JSON object from the template query, not a string.
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-08 | pass→pass | 7,218 | 2,263 | -69% | 1 | 1 | 0% | 1,259 | 1,422 | +13% | 0 | 0 | — |
case-01 | fail→fail | 14,199 | 6,420 | -55% | 1 | 1 | 0% | 2,387 | 1,442 | -40% | 0 | 0 | — |
case-02 | fail→fail | 9,134 | 4,654 | -49% | 1 | 1 | 0% | 1,479 | 1,196 | -19% | 0 | 0 | — |
case-03 | fail→fail | 7,260 | 5,809 | -20% | 1 | 1 | 0% | 1,402 | 1,230 | -12% | 0 | 0 | — |
case-04 | fail→fail | 10,501 | 8,282 | -21% | 1 | 1 | 0% | 1,753 | 1,749 | -0% | 0 | 0 | — |
case-05 | fail→pass | 6,410 | 2,016 | -69% | 1 | 1 | 0% | 1,142 | 1,325 | +16% | 0 | 0 | — |
case-06 | pass→pass | 8,233 | 2,661 | -68% | 1 | 1 | 0% | 1,518 | 1,492 | -2% | 0 | 0 | — |
case-07 | fail→pass | 14,094 | 3,311 | -77% | 1 | 1 | 0% | 2,473 | 1,424 | -42% | 0 | 0 | — |
case-22 | fail→pass | 8,828 | 4,608 | -48% | 1 | 1 | 0% | 1,512 | 1,785 | +18% | 0 | 0 | — |
case-09 | pass→pass | 10,627 | 4,048 | -62% | 1 | 1 | 0% | 1,846 | 1,503 | -19% | 0 | 0 | — |
case-10 | fail→fail | 7,569 | 1,526 | -80% | 1 | 1 | 0% | 1,160 | 1,243 | +7% | 0 | 0 | — |
case-11 | pass→pass | 7,149 | 2,592 | -64% | 1 | 1 | 0% | 1,020 | 1,418 | +39% | 0 | 0 | — |
case-12 | fail→pass | 25,252 | 6,234 | -75% | 1 | 1 | 0% | 2,520 | 2,142 | -15% | 0 | 0 | — |
case-13 | pass→pass | 6,895 | 2,102 | -70% | 1 | 1 | 0% | 1,181 | 1,282 | +9% | 0 | 0 | — |
case-14 | pass→pass | 7,586 | 2,654 | -65% | 1 | 1 | 0% | 1,245 | 1,491 | +20% | 0 | 0 | — |
case-15 | fail→pass | 9,784 | 3,302 | -66% | 1 | 1 | 0% | 1,714 | 1,596 | -7% | 0 | 0 | — |
case-16 | fail→pass | 4,655 | 2,178 | -53% | 1 | 1 | 0% | 811 | 1,258 | +55% | 0 | 0 | — |
case-17 | fail→pass | 10,781 | 2,653 | -75% | 1 | 1 | 0% | 1,772 | 1,403 | -21% | 0 | 0 | — |
case-18 | fail→pass | 6,346 | 2,002 | -68% | 1 | 1 | 0% | 1,098 | 1,287 | +17% | 0 | 0 | — |
case-19 | fail→fail | 12,102 | 2,169 | -82% | 1 | 1 | 0% | 1,862 | 1,396 | -25% | 0 | 0 | — |
case-20 | pass→pass | 6,436 | 3,458 | -46% | 1 | 1 | 0% | 1,143 | 1,625 | +42% | 0 | 0 | — |
case-21 | pass→fail | 7,182 | 6,793 | -5% | 1 | 1 | 0% | 1,140 | 1,305 | +14% | 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, and 18 counted toward the lift figure. The other 4 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 +32 percentage points is the difference between those two pass rates over the 18 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.