zero-token / README.md
ilanguage's picture
Upload README.md with huggingface_hub
f47fc70 verified
|
raw
history blame
3.58 kB

Zero Token

License: MIT OpenClaw Skill I-Lang Protocol

๐ŸŒ ็ฎ€ไฝ“ไธญๆ–‡

Zero cost. Premium output. Never offline.

Your token ran out. DeepSeek API is down. Your agent is mid-task. What now?

Zero Token gives your agent two things:

  1. A personality โ€” the same ::GENE{} behavioral DNA from Poor Man's Opus
  2. A safety net โ€” automatic fallback through free LLM providers when the main model fails

One install. One setup command. Your agent survives API outages, token exhaustion, and bill shock.


Why this exists

Poor Man's Opus makes DeepSeek output like Opus at 3% cost. Zero Token makes ANY model output premium โ€” for free. Use it as your entry point, your backup, or both.


Install & activate

# 1. Install the skill
openclaw skills install zero-token

# 2. Run setup
bash ~/.openclaw/workspace/skills/zero-token/scripts/setup.sh

# 3. Get at least one free API key (30 seconds):
#    - Gemini: https://aistudio.google.com/apikey
#    - Groq: https://console.groq.com/keys
#    - Cerebras: https://cloud.cerebras.ai

# 4. Add the key in Free-Way web console:
#    http://localhost:8787 โ†’ API Keys tab

# 5. Restart. Your agent has a free safety net.

How it works

Your prompt
    โ”‚
    โ–ผ
OpenClaw tries your main model (DeepSeek V4 Pro)
    โ”‚
    โ”œโ”€โ”€ โœ… Working โ†’ Normal response
    โ”‚
    โ””โ”€โ”€ โŒ Down / No key / Tokens exhausted
            โ”‚
            โ–ผ
        Free-Way Gateway (localhost:8787)
            โ”‚
            โ”œโ”€โ”€ Groq (Llama 3.3 70B)
            โ”œโ”€โ”€ Gemini Flash 2.5
            โ”œโ”€โ”€ Cerebras
            โ”œโ”€โ”€ OpenRouter
            โ”œโ”€โ”€ Cloudflare
            โ””โ”€โ”€ ... 13+ providers
            โ”‚
            โ–ผ
        Response โ€” same ::GENE{} DNA applied

Free model lineup

Provider Free daily limit Best model Signup time
Gemini Flash 1,500 req Gemini 2.5 Flash 30 sec
Groq 1,000 req Llama 3.3 70B 30 sec
Cerebras 1,700 req Llama 3.3 70B 30 sec
OpenRouter 200 req Multiple 30 sec
Cloudflare 10k neurons Llama 3.1 8B 1 min
Mistral 1 req/sec Mistral Small 30 sec
Cohere 20 req/min Command R+ 30 sec
NVIDIA NIM Free tier Llama 3.3 70B 1 min

When to use what

Scenario Use
First time trying agent skills Zero Token โ€” free, instant
Main model daily production Poor Man's Opus โ€” DeepSeek V4 Pro, $1.74/M
Backup when API is down Zero Token โ€” auto-switches to free models
Max output quality Poor Man's Opus โ€” full behavioral genome

Upgrade path

Zero Token has the same behavioral DNA. When you want the best:

openclaw skills install poor-mans-opus

DeepSeek V4 Pro + full SOUL genome = Opus output at 3% cost.


What's inside

  • SOUL.md โ€” 7-gene behavioral profile (ClawScan-safe)
  • SKILL.md โ€” Skill manifest
  • scripts/setup.sh โ€” One-command Free-Way gateway installer

Links

MIT License