# Launch Cloud Claw Agents

Cloud Claw deployments require correct agent choices, names, tokens, and quota checks. This skill guides new launches through the local AltLLM CLI.

## Install

```bash
npx skillstore add internet-court/cloud-claw-launch-agent
```

## Metadata

- Status: approved
- Slug: internet-court-cloud-claw-launch-agent
- Skillstore revision: r1
- Version status: missing
- Tree hash: bba9fd4d8301f3669ef79b5305dfab7050b51b59fef393637c6a52a1f1fa7513
- Author: internet-court
- GitHub username: internet-court
- License: MIT
- Repository: https://github.com/internet-court/internet-court-skill/tree/main/vendored/altlayer/cloud-claw-launch-agent
- Ref: 3f6e026a3363e0954ede7bef0cfe88d4475de137
- Supported tools: Claude, Codex, Claude Code
- Audit status: complete
- Agent install advisory: allowed
- Manual install advisory: allowed
- Artifact signature: available
- Audit attestation: unavailable
- Human verification: not\_verified
- Risk factors: external\_commands, filesystem
- Quality score: 74
- Public page: https://skillstore.pages.dev/skills/internet-court-cloud-claw-launch-agent
- Manifest: https://skillstore.pages.dev/api/skills/internet-court-cloud-claw-launch-agent/manifest

## Capabilities

- Guides only the new Cloud Claw deployment workflow.
- Maps OpenClaw, PicoClaw, and Ottie to internal agent type values.
- Applies name, agent type, model, and Telegram token constraints.
- Surfaces quota, free slot, credit, and payment prerequisites before launch.
- Points users to expected launch responses and common failure states.

## Use Cases

- Prepare a New OpenClaw Deployment: Set the deployment name, model, token source, and allowed users before launching an OpenClaw agent.
- Launch a Private Telegram Agent: Collect Telegram token inputs and allowed user IDs for a PicoClaw or Ottie launch.
- Check Account Readiness: Review quota, credits, payments, and free-slot status before creating a Cloud Claw deployment.

## Prompt Templates

### Prepare First Launch

```
I want to launch my first Cloud Claw deployment. Ask me for the required fields and explain any missing prerequisites.
```

### Launch PicoClaw Bot

```
For a PicoClaw launch, collect the deployment name, Telegram token source, and allowed Telegram users. Confirm before running any command.
```

### Configure OpenClaw Model

```
Help me choose an OpenClaw model. Check the allowed model options and prepare the launch parameters for review.
```

### Run Controlled Production Launch

```
Plan a controlled Cloud Claw rollout for a paid account. Verify quota, credits, agent type, secret handling, and rollback boundaries before deployment.
```

## Limitations

- Does not start, stop, delete, or inspect existing deployments.
- Requires the local AltLLM CLI, authentication, and account access.
- Cannot bypass quota, credits, free slots, or payment requirements.
- Requires users to provide valid secrets through safe local inputs.

## Best Practices

- Confirm deployment name, agent type, model, and cost impact before running the launch command.
- Use environment variables or secure files for tokens, and redact secrets from chat history.
- Check account quota, credits, and free-slot status before starting a launch.

## Anti Patterns

- Running a deployment before the user confirms final parameters.
- Pasting raw Telegram or API tokens directly into chat.
- Using this skill to manage existing deployments after launch.

## Security Audit

- Audited at: 2026-07-21T02:57:45.638\+00:00
- Summary: All 19 static findings are false positives. Backticks are Markdown formatting rather than executable shell or Ruby code, and the relative paths are fixed documentation references. No prompt injection, credential exfiltration, or untrusted command construction was found in the reviewed files.

## Stats

- Views: 0
- Downloads: 4
- Favorites: 1
- Popularity score: 0