Pricing
Free, with every feature unlocked. Bring your own model key or run one locally.
- Every feature unlocked
- 24 tools, 35 skills, Telegram / Slack / Signal
- Your own API key, a local model, or a Claude/ChatGPT subscription
- Cross-device sync + encrypted relay
- Secure gateway from any device
- Unlimited local model usage
- Bring your own API keys (all models)
- Community support
Need an Enterprise dedicated runtime, VPC mode, or per-org isolation? .
Bring your own keys (BYOK)
This is how Foxl reaches a model. Add your Anthropic, OpenAI, or Google API key in Settings and all requests route directly to the provider - no markup, and you pay the provider directly at their published rates. Every model is available this way, Opus included. You can also use an existing Claude or ChatGPT subscription.
Local models are always free
Foxl detects Ollama automatically. Run Llama, Mistral, Qwen, or any GGUF model locally with zero cost, zero API keys, zero internet. Conversations stay entirely on your machine.
What's included in every plan
- Full agent with 24 built-in tools (browser, terminal, file system, code search, memory, etc.)
- 35 installable skills (meeting transcription, email triage, code review, research, etc.)
- Reach it from Telegram, Slack or Signal, plus Outlook and Slack integrations
- Cron scheduler for automated agent tasks
- Encrypted relay for cross-device access (phone, tablet, any browser)
- Multi-agent orchestration with parallel subagents
- Real browser control with your existing logins
- All data stays on your device. No cloud storage.
FAQ
Is it really free?
Yes. Every feature is unlocked and there is nothing to buy. Foxl does not supply model inference, so you reach a model with your own provider key, a model you run locally, or an existing Claude or ChatGPT subscription - and you pay that provider directly, not us.
Do I need a credit card?
No. Download, sign up with email, and start using the agent immediately.