Use Obscurify.ai as a model provider in OpenClaw

OpenClaw talks to any OpenAI-compatible API, so you can point it at Obscurify.ai with a few lines of config. You get 200+ models through one key and an account number instead of an email.

Tested on 2026-10-31 with OpenClaw 2026.10.9 against the live API, including an agent turn that wrote a file through a tool.

Before you start

You need:

  • OpenClaw, on Node 24.22.1 or newer: npm install -g openclaw@latest
  • An Obscurify.ai API key. Create an account, add credit on the account page, then click Show API Key. New accounts may get $1 of starter credit while this month’s promo budget lasts.

Put the key in your environment. It starts with sk-obs1_:

export OBSCURIFY_API_KEY="sk-obs1_..."

Add the provider

Run this once. OpenClaw saves it to ~/.openclaw/openclaw.json, and it keeps the variable’s name rather than your key:

openclaw config set models.providers.obscurify '{
  "baseUrl": "https://obscurify.ai/v1",
  "apiKey": "${OBSCURIFY_API_KEY}",
  "api": "openai-completions",
  "models": [
    { "id": "deepseek/deepseek-v4-flash", "name": "DeepSeek V4 Flash", "reasoning": true, "contextWindow": 1048576, "maxTokens": 16000, "compat": { "supportsUsageInStreaming": true } },
    { "id": "z-ai/glm-5.3", "name": "GLM 5.3", "reasoning": true, "contextWindow": 1048576, "maxTokens": 16000, "compat": { "supportsUsageInStreaming": true } },
    { "id": "qwen/qwen3-coder", "name": "Qwen3 Coder 480B", "reasoning": false, "contextWindow": 262000, "maxTokens": 16000, "compat": { "supportsUsageInStreaming": true } }
  ]
}' --strict-json --merge
openclaw models set obscurify/deepseek/deepseek-v4-flash

compat.supportsUsageInStreaming makes OpenClaw ask for token counts in streams, so it can track context size.

Try it

openclaw models list --provider obscurify --plain
openclaw agent --local --agent main --message "Reply with exactly: pong"

Then give it something that needs a tool:

openclaw agent --local --agent main --message "Create a file named hello.txt in the current directory containing exactly: obscurify works. Then reply DONE."

Use other models

Add any model to the models list. Models are named obscurify/<id>, and the ID keeps its own prefix, as in obscurify/z-ai/glm-5.3. The full list is at /v1/models, with prices on the model list.

Or install the plugin

The plugin does the same job and keeps the whole model list up to date on its own:

openclaw plugins install clawhub:@obscurify/openclaw-provider
openclaw onboard --auth-choice obscurify-api-key --obscurify-api-key "$OBSCURIFY_API_KEY"

What it costs

Each request is charged from your balance at the model’s price. Agents use far more tokens than chat, because each step resends the system prompt, tools and history: small tasks in our tests averaged about 20,000 tokens. Pay As You Go stops at 1M tokens a month, Casual at 10M, and Pro has no limit. See pricing.

If something fails

  • 402: there’s no balance on the account. Add credit on the account page.
  • 403: your plan doesn’t include that model. Frontier models need Casual or Pro, and uncensored models need Pro.
  • 404 model_not_found: check the ID against /v1/models, and keep its prefix.
  • 429: you’re going too fast, or you’ve hit the plan’s monthly token limit.
  • “returned no token usage” in the logs: add compat.supportsUsageInStreaming to that model, as above.

What stays private

Prompts and responses aren’t stored on Obscurify.ai’s servers, and text requests only use zero-data-retention routes, which rely on providers’ published policies. Providers don’t see who sent a prompt, but they do see its text, and that includes any files your agent reads and sends. Details are in the privacy policy.

More agents: Obscurify.ai for agents.