메뉴
Enable custom model support — use local models (OMLX, Ollama) or any third-party model endpoint (HuggingFace, Together AI, Groq, etc.) alongside your primary agent SDK. Triggers on "custom models", "local models", "add custom", "add local".
Codex 또는 Claude로 설치 이 Prompt를 복사해 Codex, Claude 또는 다른 어시스턴트에 붙여 넣으면 Skill 페이지를 검토하고 설치를 진행할 수 있습니다.
SOC 직업 분류 기준
Register an email mailbox for skills to operate on — reading, triage, drafting, bulk archive. NOT a channel; inbound emails will not wake the agent. For email-as-a-channel, use /add-gmail instead. Supports Google Workspace (Gmail), Microsoft 365 (Outlook/Exchange), and IMAP.
Set up email archive system — interactive calibration builds sender rules, creates folder taxonomy, and configures batch processing to reach inbox zero. Requires at least one email account registered via /add-email-account. Run this before using /email-archive in your main chat.
Batch-classify old emails into organized folders, working toward inbox zero. Uses sender rules (free, fast) with LLM fallback for unknown senders. Supports interactive, supervised, and autonomous modes. Provider-agnostic — works with any email account configured in config.yaml. Use /email-archive to process a batch, /email-archive status for progress.
Add a model endpoint — local (OMLX, Ollama) or remote (HuggingFace, Together AI, Groq, or any OpenAI-compatible URL). Triggers on "add endpoint", "add model", "model endpoint", "connect model".
Add Google Gemini agent SDK. Free tier available (60 req/min). Built-in tools, MCP support, 11 lifecycle hooks. Authenticates via API key.
Run initial NanoClaw setup. Use when user wants to install dependencies, authenticate messaging channels, register their main channel, or start the background services. Triggers on "setup", "install", "configure nanoclaw", or first-time setup requests.
| name | add-custom-models |
| description | Enable custom model support — use local models (OMLX, Ollama) or any third-party model endpoint (HuggingFace, Together AI, Groq, etc.) alongside your primary agent SDK. Triggers on "custom models", "local models", "add custom", "add local". |
Enable support for custom model endpoints — local models, third-party providers, or self-hosted deployments. Uses the Codex SDK as the bridge to any OpenAI-compatible endpoint.
ls src/runtime/codex-runtime.ts 2>/dev/null && echo "CODEX_INSTALLED" || echo "CODEX_MISSING"
If CODEX_MISSING, the Codex SDK is required as the bridge for custom models. Invoke /add-agentSDK-codex first, then return here.
grep -q 'OMLX_URL\|LITELLM_URL' .env 2>/dev/null && echo "ENDPOINTS_CONFIGURED" || echo "NO_ENDPOINTS"
Custom models let you use any OpenAI-compatible endpoint alongside your primary agent SDK.
This works by routing through the Codex SDK — when you select a custom model via
/model, the system automatically uses Codex with the custom endpoint URL.Common endpoints: • OMLX — local models optimized for Apple Silicon (persistent KV cache, tool calling) • Ollama — easy local model hosting • Together AI, Groq, Fireworks — fast third-party inference • HuggingFace Inference — hosted open-source models • Self-hosted vLLM, llama.cpp — your own deployment
Run
/add-model-endpointto add endpoints one at a time.
After running /add-model-endpoint for each desired provider, verify models appear in /model command output.
Only needed if this is the first time installing custom model support:
npm run build && ./container/build.sh
launchctl kickstart -k gui/$(id -u)/com.nanoclaw # macOS
# Linux: systemctl --user restart nanoclaw
Send /model in Telegram. You should see your custom models listed below the cloud models.
Switch to a custom model:
/model omlx/llama-3.1-8b
Send a message and verify it responds using the local/custom model.
Switch back:
/model gpt-5.4-mini
Remove endpoint URLs from .env and remove the custom model entries from AVAILABLE_MODELS in src/config.ts. Rebuild and restart. The Codex SDK can remain installed.