# Host AI Workflow Runner with Echo: Stored Input and Output Artifacts

## Summary
Deploy an AI workflow runner built with Echo on Ample using the generated downloads pattern. Compute runs the app in an isolated microVM behind a public HTTPS URL, a managed PostgreSQL 16 database is auto-provisioned and injected as DATABASE_URL, and a private S3-compatible bucket holds objects with credentials delivered as encrypted environment variables. Verified on Echo: an export query rendered to a CSV artifact stored with metadata in a private bucket and returned only to an authorized caller (401 without token, export=ok rows=3). Not separately tested: your export formats, retention policy, and entitlement model; treat the AI workflow runner-specific behavior as your application code.

## Representative Queries
- Host AI workflow runner with Echo: stored input and output artifacts
- Where can I host AI workflow runner built with Echo?
- I need an input/output artifact contract, object metadata, and access boundaries.

## Resource Requirements
- primitive:compute
- primitive:postgres
- primitive:s3-compatible-object-storage

## Infrastructure Requirements
- **Compute**
  - Status: verified
  - Summary: Apps run in isolated x86_64 Firecracker microVMs that auto-pause when idle and wake on request; sizes are the priced VM sizes.
  - [More Info](/content/discover/primitive/compute/index.html)

- **Postgres**
  - Status: verified
  - Summary: Managed PostgreSQL 16 runs in its own microVM and is auto-provisioned when an app needs a database and no DATABASE_URL is supplied.
  - [More Info](/content/discover/primitive/postgres/index.html)

- **S3-compatible Object Storage**
  - Status: verified
  - Summary: Buckets are S3-compatible with issued credentials; PutObject and GetObject are verified by canary. Other S3 operations are not verified.
  - [More Info](/content/discover/primitive/s3-compatible-object-storage/index.html)

## Prerequisites
1. An Echo project that builds and starts with the documented commands (CGO_ENABLED=0 go build -o app ./... then ./app on the ubuntu-24.04 template (go.sum committed for a reproducible build))
2. A PostgreSQL driver reading DATABASE_URL at runtime (auto-provisioned when omitted, or supplied with --env)
3. A bucket from `ample bucket create` with its credentials passed as encrypted S3_* environment variables
4. An Ample account token with servers:write, databases:read, buckets:read

## Workflow Steps
1. **Build and Start**
   - CGO_ENABLED=0 go build -o app ./... then ./app on the ubuntu-24.04 template (go.sum committed for a reproducible build). The server must bind 0.0.0.0 on PORT.

2. **Implement the Pattern on PostgreSQL**
   - The fixture's module implements generated downloads: an export query rendered to a CSV artifact stored with metadata in a private bucket and returned only to an authorized caller (401 without token, export=ok rows=3). Copy the approach into your schema; keep migrations idempotent and run them with --release-command.

3. **Wire Object Storage**
   - Create the bucket(s), then pass endpoint, region, bucket and keys as --env values. Use path-style addressing. Keep private data in an unpublished bucket.

4. **Deploy**
   - Run the synchronous deploy once and read the result (exit 0 live, 1 failed, 2 blocked). Re-running with no change is a no-op.
   - **Command:** ample deploy . --name  --public --env S3_ENDPOINT=... --env S3_REGION=... --env S3_BUCKET=... --env S3_ACCESS_KEY_ID=... --env S3_SECRET_ACCESS_KEY=...

5. **Verify**
   - Fetch the live URL and the pattern self-test route(s) (/p/generated-downloads) from the example; then run your own checks. On failure read `ample logs  --kind build` then `--kind runtime`.
   - **Command:** ample logs  --kind build

## Next Actions
- **[Browse the catalog index](https://api.ample.computer/v1/catalog)**
- **Search published recipes by intent, stack and constraints**
- **Prepare a side-effect-free deployment plan for an authorized project**
- **Read the existing agent authentication setup**
- **Browse AI workflow runner**
- **Browse Echo**
- **Browse Generated downloads**
- **Browse Deploy AI app**

## Limitations
- Verified on the ubuntu-24.04 template at s-1vcpu-1gb with the example fixture; other sizes, templates, and Echo major versions are not verified.  
- The AI workflow runner itself (your export formats, retention policy, and entitlement model) is application code and was not separately tested; the pattern checks are what was verified.
- Region, compliance attestations, and request-duration limits are unknown and not claimed.  
- Apps auto-pause when idle and wake on the next request; always-on is an operator setting, not a plan feature.