@init-computer/cli

Init CLI

Official CLI for Init. Generate, deploy, and run backend endpoints from your terminal in seconds.

Init is a programmable backend platform: describe the capability you need, and Init creates and operates the API, data, storage, and runtime behind it.

The init command brings Init to terminal workflows, scripts, CI jobs, and local debugging. Use it to generate deployed backend endpoints, run them, edit versions, manage resources, and inspect available packages.

Website: init.computer Documentation: Init.computer

Project suites

init project generate "Build ticket intake and status lookup" --name "Customer Support"
init project edit <project-id> "Add ticket assignment"

Install

npm install -g @init-computer/cli

Get an API key

Create an account at init.computer to get an API key.

Then log in with the CLI:

init login --api-key init_sk_...

Or paste your API key interactively:

init login

The CLI stores credentials in ~/.init/config.json. Treat that file like a secret.

Check or clear auth:

init auth status
init logout

For CI, scripts, and server environments, use INIT_API_KEY instead:

export INIT_API_KEY=init_sk_...

For local development only, the CLI also resolves INIT_API_KEY from a .env file in the current directory after an explicit flag and the process environment. It reads that one key only; it never uploads, lists, or synchronizes .env files.

Generate an endpoint

Use init generate to create a deployed backend endpoint from a prompt.

init generate "Create a lead enrichment API. Input fields: companyName, website, targetPersona. Output fields: companySummary, buyingSignals, recommendedPersonas, nextSteps. Research the company and return structured JSON only."

Generate from a JSON request body when the prompt is long:

init generate --body request.json

Example request.json:

{
  "prompt": "Create a lead enrichment API for B2B sales teams. Input fields: companyName string required, website url required, targetPersona string optional. Output fields: companySummary string, buyingSignals array with title evidence urgency, recommendedPersonas array with role reason outreachAngle, nextSteps array of strings. Store each generated brief and return structured JSON only.",
  "projects": ["Sales"]
}

The response includes the live url, route, resourceId, versionId, input schema, output schema, and generated pipeline.

Save the resourceId if you want to inspect, edit, organize, automate, or version the endpoint later.

Run an endpoint

Use init run to call a generated endpoint.

Run by route:

init run abc123 --data companyName=Snowflake --data website=https://www.snowflake.com --data targetPersona="VP Sales"

Run by full URL:

init run https://api.init.computer/v1/abc123 --body input.json

Read JSON from stdin:

cat input.json | init run abc123 --body -

Use route or url to run the deployed endpoint. Use resourceId for management tasks such as edits and versions.

Edit an endpoint

Edit an existing endpoint by passing its resourceId.

init generate "Keep the same behavior, but add csv and confidenceScore outputs." --resource-id resource_uuid

Init keeps the same resource and creates a new version.

Work with files

Generated endpoints can accept file input when the endpoint needs uploaded files.

init run quarterly-report-reader \
  --file file=report.pdf \
  --data summaryStyle=executive

You can also send multipart input from a request file:

init run quarterly-report-reader --body multipart-request.json

Manage resources

Resources are the APIs you have generated.

init resources
init resources --q lead --status enabled
init resource resource_uuid
init resource resource_uuid versions
init resource resource_uuid activate version_uuid
init resource resource_uuid update --body patch.json

Example patch.json:

{
  "name": "Lead Enrichment",
  "status": "enabled"
}

Discover packages

Packages are the capabilities available to generated endpoints.

A package can be an Init-provided capability, an external app, or one of your own tools. Packages can give endpoints access to compute, storage, databases, models, APIs, and integrations such as OpenAI, Anthropic, Neon, Supabase, or custom internal services.

Use packages to see what your generated endpoints can connect to and use during execution.

init packages
init packages --category Database
init packages --category Model
init packages --category Integration
init packages --enabled-category Storage

For example, you can generate an endpoint that uses connected packages:

init generate "Create a customer support triage API. Input fields: message and customerId. Output fields: category, priority, summary, suggestedReply, and shouldEscalate. Use the connected Anthropic package to classify and summarize the request. Use the connected Supabase package to store the support ticket. Return structured JSON only."

Once your own tools are connected to Init as packages, generated endpoints can call them the same way they call built-in or third-party capabilities.

Output

By default, the CLI prints pretty JSON.

init resources

Use compact JSON for scripts:

init resources --json

Errors

The CLI prints failed requests as JSON with success: false, error, and status.

• 402: Plan limit reached. Upgrade your plan or enable overages.
• 429: Too many concurrent requests. Retry after current runs finish.
• 500: Internal server error. The public response is intentionally broad.
• 503: Init is temporarily unavailable.

Global options

--api-key <key>       Use an API key without saving it
--base-url <url>      API origin, defaults to https://api.init.computer
--json                Print compact JSON
--pretty              Print pretty JSON

Environment

INIT_API_KEY       API key for CI or server usage
INIT_API_BASE_URL  API origin override
INIT_CONFIG_DIR    Saved auth config directory, defaults to ~/.init

Documentation

For full documentation, guides, and API reference, visit Init.computer.