How AI Detection Works

When you import a repository, FluxNow scans it to detect everything needed to build and deploy your application. Here is what it looks for and how it decides.

Runtime detection

FluxNow reads standard dependency and project files:

File	Detected runtime
package.json with bun.lock, bun.lockb, or bun engine	Bun
package.json (no Bun lock)	Node.js
requirements.txt, pyproject.toml, Pipfile	Python
go.mod	Go
Gemfile	Ruby
Dockerfile	Uses your Dockerfile directly

Port detection

FluxNow checks in order:

1. EXPOSE instruction in a Dockerfile
2. Framework-specific defaults (e.g., Next.js → 3000, FastAPI → 8000, Rails → 3000)
3. Start scripts in package.json for explicit --port or -p flags

Service detection

FluxNow looks at environment variable patterns used by your application:

Pattern	Provisioned service
DATABASE_URL, POSTGRES_*, PG_*	Postgres
REDIS_URL, REDIS_HOST	Redis
S3_BUCKET, S3_ENDPOINT, AWS_*	S3-compatible storage

Build strategy

FluxNow selects a build strategy in this order:

1. Dockerfile — if a Dockerfile is present at the repo root
2. Railpack — if no Dockerfile is found, Railpack auto-builds a container from the detected runtime
3. Auto — fallback when no other signal is available

What you can override

After detection, the review panel lets you adjust any setting before deploying:

• Runtime and version
• Exposed port
• Which services to enable or disable
• Environment variable names and placeholders

After the initial deploy, all settings are managed in the project dashboard. There is no configuration file added to your repository.