setup

Included with Lifetime

$97 forever

Configure harness optimize-subsystem hardware targets — Apple Silicon (MLX), NVIDIA server (CUDA), or RunPod cloud GPU. One-time setup stored persistently. Use when user mentions hardware setup, SSH targets, GPU configuration, or says 'configure optimization targets'.

Cloud & DevOps

What this skill does


# Harness: Optimize Setup

Guide the user through configuring hardware targets for the optimization loop. This is the ONE interactive entry point for the optimize subsystem — everything else is headless.

Config persists across sessions at `${CLAUDE_PLUGIN_DATA}/config.json` (fallback: `~/.harness/config.json`).

## Step 1: Check existing config

```bash
cat "${CLAUDE_PLUGIN_DATA:-${HOME}/.harness}/config.json" 2>/dev/null || echo "NO_CONFIG"
```

If config exists, show current targets and ask if the user wants to update or start fresh.

## Step 2: Auto-detect local hardware

```bash
bash "${CLAUDE_PLUGIN_ROOT}/scripts/detect-hardware.sh"
```

This detects Apple Silicon (MLX), NVIDIA GPU (CUDA), or CPU-only.

## Step 3: Configure targets

### Local

If Apple Silicon or NVIDIA GPU detected, ask for the path to an experiment repo (or any project to optimize). Validate:

```bash
[ -d "<path>" ] && echo "VALID" || echo "INVALID"
```

For ML training repos, also check:
```bash
[ -f "<path>/program.md" ] && [ -f "<path>/train.py" ] && echo "ML_REPO" || echo "GENERIC"
```

### Remote NVIDIA Server

Ask if they have an SSH-accessible NVIDIA GPU server. If yes:

```bash
bash "${CLAUDE_PLUGIN_ROOT}/scripts/test-ssh.sh" "<ssh_host>"
```

Collect:
- SSH hostname (e.g., `ml-server`, `[email protected]`)
- Remote working directory

Identify the installed fork:
```bash
ssh -o ConnectTimeout=3 "<ssh_host>" "cd <path> && git remote get-url origin 2>/dev/null && head -5 train.py" 2>/dev/null
```

Repo recommendations:
- Consumer NVIDIA (RTX 20/30/40/50): [flight505/autoresearch-blackwell](https://github.com/flight505/autoresearch-blackwell)
- Datacenter (H100, A100): [karpathy/autoresearch](https://github.com/karpathy/autoresearch)

### RunPod (Cloud GPU)

Ask if the user wants RunPod cloud GPU access. If yes:
- Collect API key. No account: "Sign up at https://runpod.io?ref=wjm4q5bw"
- Pod provisioning is manual for now; API key stored for future automation.

## Step 4: Clone experiment target (optional)

If the user doesn't have an experiment repo yet, offer to clone one:

```bash
bash "${CLAUDE_PLUGIN_ROOT}/scripts/clone-target.sh" "<hardware>" "<destination>"
```

This clones the appropriate autoresearch fork based on hardware.

## Step 5: Write config

```bash
cat << 'CONFIGEOF' | bash "${CLAUDE_PLUGIN_ROOT}/scripts/write-config.sh"
{
  "version": 1,
  "targets": {
    "local": {
      "enabled": true/false,
      "path": "<path>",
      "backend": "mlx" | "cuda" | "cpu",
      "description": "<auto-detected hardware>"
    },
    "server": {
      "enabled": true/false,
      "ssh_host": "<hostname>",
      "path": "<remote path>",
      "backend": "cuda",
      "gpu_type": "<detected GPU>",
      "repo": "<git remote URL>",
      "description": "<GPU name + fork>"
    },
    "runpod": {
      "enabled": true/false,
      "api_key": "<key>",
      "gpu_type": "NVIDIA RTX 4090",
      "description": "RunPod cloud GPU"
    }
  }
}
CONFIGEOF
```

## Step 6: Verify

Read back and summarize:
```bash
cat "${CLAUDE_PLUGIN_DATA:-${HOME}/.harness}/config.json"
```

Then explain next steps:
- The orchestrator can now use `loop` phases in workflows
- Direct agent spawn: the optimizer agent reads this config for server targets
- The advisor agent can analyze projects against these targets

Files: 2

Size: 3.6 KB

Complexity: 14/100

Category: Cloud & DevOps

Source: https://github.com/flight505/flight505-marketplace/tree/main/harness/skills/setup

Related in Cloud & DevOps

appbuilder-action-scaffolder

Included

Create, implement, deploy, and debug Adobe Runtime actions with consistent layout, validation, and error handling. Use this skill whenever the user needs to add actions to an App Builder project, understand action structure (params, response format, web/raw actions), configure actions in the manifest, use App Builder SDKs (State, Files, Events, database), deploy and invoke actions via CLI, debug action issues, or implement patterns such as webhook receivers, custom event providers, journaling consumers, large payload redirects, action sequence pipelines, and Asset Compute workers. Also trigger when users mention serverless functions in Adobe context, action logging, IMS authentication for actions, or cron-style scheduled actions.

Cloud & DevOpsscripts

orchestrating-datacloud

Included

Salesforce Data Cloud product orchestrator for connect→prepare→harmonize→segment→act workflows. Use this skill when the user needs a multi-step Data Cloud pipeline, cross-phase troubleshooting, or data space and data kit management. TRIGGER when: user needs a multi-step Data Cloud pipeline, asks to set up or troubleshoot Data Cloud across phases, manages data spaces or data kits, or wants a cross-phase sf data360 workflow. DO NOT TRIGGER when: work is isolated to a single phase (use the matching phase-specific skill), the task is STDM/session tracing/parquet telemetry (use observing-agentforce), standard CRM SOQL (use querying-soql), or Apex implementation (use generating-apex).

Cloud & DevOpsscripts

github-project-automation

Included

Automate GitHub repository setup with CI/CD workflows, issue templates, Dependabot, and CodeQL security scanning. Includes 12 production-tested workflows and prevents 18 errors: YAML syntax, action pinning, and configuration. Use when: setting up GitHub Actions CI/CD, creating issue/PR templates, enabling Dependabot or CodeQL scanning, deploying to Cloudflare Workers, implementing matrix testing, or troubleshooting YAML indentation, action version pinning, secrets syntax, runner versions, or CodeQL configuration. Keywords: github actions, github workflow, ci/cd, issue templates, pull request templates, dependabot, codeql, security scanning, yaml syntax, github automation, repository setup, workflow templates, github actions matrix, secrets management, branch protection, codeowners, github projects, continuous integration, continuous deployment, workflow syntax error, action version pinning, runner version, github context, yaml indentation error

Cloud & DevOpsscripts

sf-datacloud

Included

Salesforce Data Cloud product orchestrator for connect→prepare→harmonize→segment→act workflows. TRIGGER when: user needs a multi-step Data Cloud pipeline, asks to set up or troubleshoot Data Cloud across phases, manages data spaces or data kits, or wants a cross-phase `sf data360` workflow. DO NOT TRIGGER when: work is isolated to a single phase (use the matching sf-datacloud-* skill), the task is STDM/session tracing/parquet telemetry (use sf-ai-agentforce-observability), standard CRM SOQL (use sf-soql), or Apex implementation (use sf-apex).

Cloud & DevOpsscripts

fabric-cli

Included

Use this skill for Fabric.so CLI workflows with the `fabric` terminal command: diagnose/install/login, search or browse a Fabric library, save notes/links/files, create folders, ask the Fabric AI assistant, manage tasks/workspaces, generate shell completion, check subscription usage, produce JSON output, and use Fabric as persistent agent memory. Do not use for Microsoft Fabric/Azure/Power BI `fab`, Daniel Miessler's Fabric framework, Python Fabric SSH, Fabric.js, or textile/fashion fabric.

Cloud & DevOpsscripts

lark

Included

Lark/Feishu CLI skills: lark-cli operations for docs, markdown, sheets, base, calendar, im, mail, task, okr, drive, wiki, slides, whiteboard, apps, approval, attendance, contact, vc, minutes, event. Use when the user needs to operate Lark/Feishu resources via lark-cli, send messages, manage documents, spreadsheets, calendars, tasks, OKRs, deploy web pages, or any Feishu/Lark workspace operations.

Cloud & DevOpsscripts