// Create a new Cloudflare app in this repo with the expected package scripts, Doppler shape, and CI workflow wiring.
| name | creating-an-app |
| description | Create a new Cloudflare app in this repo with the expected package scripts, Doppler shape, and CI workflow wiring. |
| publish | false |
Use this when adding a new Cloudflare app under apps/.
Keep the app contract small:
alchemy:upalchemy:downtest:e2e if the app has live preview testsThe package scripts should own only the app action. Doppler selection belongs outside the app script.
Cloudflare deploy workflows are generated from .github/ts-workflows/.
The current pattern is:
packages/shared/src/apps/new-style-cloudflare-apps.ts.scripts/preview/apps.ts..github/ts-workflows/workflows/deploy-<app>.ts.pnpm -C .github/ts-workflows generate.The shared app manifest is the source of truth for:
The preview registry adds:
Preview deploys do not live in app-local routers anymore. They run through the repo preview router:
doppler run --project _shared --config prd -- pnpm preview sync --pull-request-number 1234
doppler run --project _shared --config prd -- pnpm preview cleanup --pull-request-number 1234
Workflow rules:
preview_Nmain deploy prdmain deploy successes post to Slackmain deploy failures post to Slackpaths include the app folder, contract folder, and shared new-style deploy paths from packages/shared/src/apps/new-style-cloudflare-apps.tsDo not add preview logic back into apps/<app>/scripts/router.ts just to satisfy CI.
Use the new-doppler-project skill for the project/config setup.
The app package should work with:
doppler run --project <app> --config preview_2 -- pnpm exec tsx ./alchemy.run.ts
doppler run --project <app> --config prd -- pnpm exec tsx ./alchemy.run.ts
Query and analyze Cloudflare Workers traces through the general Cloudflare API MCP server. Use when diagnosing Workers traces, distributed tracing spans, observability events, subrequest chains, or when an agent needs Cloudflare trace data programmatically.
Debug MCP servers with the official MCP Inspector UI and CLI. Use when an MCP server will not connect, auth fails, tools/resources/prompts are missing, or a tool call needs to be reproduced outside Codex/Claude.
Drizzle ORM database migration workflow for packages that still use Drizzle (agents, example). Do NOT use for apps/os — that app uses sqlfu/D1.
Monitoring workflow for architect: Cloudflare Worker checks, task updates, and Slack escalation.
Debug failures in the Cloudflare OS worker using the Cloudflare MCP server. Use when diagnosing 500s, missing logs, or request-level regressions in production or preview.
PostHog alert responder workflow with required Slack deep-links and escalation rules.