ppq-private-mode

name	ppq-private-mode
description	Configure PPQ.AI Private Mode encrypted models in OpenClaw
user-invocable	true

PPQ.AI Private Mode -- Configure Encrypted Models

This skill installs and configures OpenClaw to use PPQ.AI's end-to-end encrypted models.

Step 1: Install the plugin

Run this command to install the PPQ Private Mode plugin:

openclaw plugins install ppq-private-mode

Step 2: Verify the plugin is installed

Check that the plugin exists:

ls ~/.openclaw/extensions/ppq-private-mode/

If the directory does not exist, the install may have failed. Try running the install command again.

Step 3: Get the user's PPQ API key

First, check if the user already has a PPQ API key configured. Read ~/.openclaw/openclaw.json and look for an existing key in any of these locations:

• plugins.entries.ppq-private-mode.config.apiKey
• Any provider config with a baseUrl pointing to api.ppq.ai

If a valid key is found (starts with sk-), reuse it -- no need to ask the user. Let them know you found their existing key.

If no key is found, ask the user for their PPQ.AI API key. They can find it at https://ppq.ai/api-docs -- it starts with sk-.

Do not proceed until you have a key.

Step 4: Add BOTH the provider AND the plugin config to OpenClaw

IMPORTANT: You must add BOTH sections below to ~/.openclaw/openclaw.json. Do not skip either one. Merge them into the existing config, preserving all existing settings.

Here is the complete config to merge (replace USER_API_KEY with the actual key):

{
  "models": {
    "providers": {
      "ppq-private": {
        "baseUrl": "http://127.0.0.1:8787/v1",
        "apiKey": "unused",
        "api": "openai-completions",
        "models": [
          { "id": "private/kimi-k3", "name": "private/kimi-k3" },
          { "id": "private/gpt-oss-120b", "name": "private/gpt-oss-120b" },
          { "id": "private/llama3-3-70b", "name": "private/llama3-3-70b" },
          {     
               
               
        
      
    
  
   
     
       
         
           
        
      
    
  

After merging, verify that ~/.openclaw/openclaw.json contains BOTH:

1. A ppq-private entry under models.providers with the 6 private models listed
2. A ppq-private-mode entry under plugins.entries with the user's API key

If either is missing, the private models will not work.

Step 5: Restart the gateway

systemctl --user restart openclaw-gateway.service

Step 6: Confirm to the user

Tell the user PPQ Private Mode is configured. Available encrypted models:

• private/kimi-k3 -- Moonshot flagship multimodal MoE, reasoning, strong coding, 256K context
• private/gpt-oss-120b -- Cost-efficient general use
• private/llama3-3-70b -- Open-source tasks
• private/glm-5-2 -- Agentic engineering with long-horizon tool use, 384K context
• private/gemma4-31b -- Vision plus text with thinking mode, 262K context
• private/deepseek-v4-flash -- Fast, cost-efficient reasoning and coding, 1M context

Switch with: openclaw models set private/kimi-k3

Troubleshooting

• "Authentication error": Check API key and account balance at https://ppq.ai
• "Attestation failed": Enclave may be temporarily unavailable. Wait and restart: systemctl --user restart openclaw-gateway.service
• "Model not allowed": The ppq-private provider is missing from models.providers in openclaw.json. Add it using the config block above.
• Port conflict on 8787: Add "port": 8788 to the plugin config in openclaw.json

About

PPQ.AI provides pay-per-query AI with no subscriptions. Private models run inside hardware secure enclaves. Learn more at https://ppq.ai