faradaiwire guard

faradaiwire guard 🛡️

The guard subcommand manages the local Small Language Model (SLM) classifiers used by FaradAiWire’s Deep-Path Neural Security Guard.

---

Syntax

faradaiwire guard <subcommand> [arguments] [flags]

---

Subcommands

Subcommand	Arguments	Description
list	None	Lists all supported SLM models, their parameter sizes, provider engines, and local installation status.
status	None	Displays real-time operational status, active model, provider endpoint, and inference latency.
install	<model-id>	Triggers automated local model installation (e.g. invokes Ollama pull or verifies Docker image).
uninstall	<model-id>	Removes local model weights to reclaim workstation disk space.

---

Examples & Output

1. List Supported Models:

faradaiwire guard list

AVAILABLE SLM MODELS:
 ID                   NAME                 PROVIDER    PARADIGM        STATUS
 prompt-guard-docker  Meta PromptGuard     docker      Classification  READY (8088)
 llama-guard3:1b      Llama Guard 3 1B     ollama      DecisionModel   READY (11434)
 qwen3guard-0.6b      Qwen3 Guard 0.6B     ollama      DecisionModel   AVAILABLE
 granite-guardian     IBM Granite 3.0      ollama      DecisionModel   AVAILABLE

2. Check Guard Status:

faradaiwire guard status

ACTIVE SLM SECURITY GUARD:
 Provider:          Ollama (http://127.0.0.1:11434)
 Active Model:      llama-guard3:1b
 Quantization:      Q4_K_M (4-bit)
 Context Size:      2048 tokens
 Average Latency:   24.5 ms
 Status:            HEALTHY

3. Install a New Guard Model:

faradaiwire guard install qwen3guard-0.6b

FaradAiWire communicates with the local Ollama API to pull and verify model weights.