Back to PacksAGENT PACK

Local AI Pack

Run AI workflows entirely on local hardware. Configure models, optimize performance, build RAG pipelines, and fine-tune — all without cloud dependencies.

For

Developers running AI models on consumer hardware, laptops, or private servers.

Outcome

An optimized local model deployment with benchmarks, a RAG pipeline, and fine-tuning validation.

Agents in this pack

Ollama Claw

Set up and optimize Ollama with model selection, GPU config, and Modelfiles

ollama-claw 2.

RAG Claw

Design RAG systems with chunking, embedding, retrieval, and generation

rag-claw 3.

Quantization Claw

Evaluate quantization strategies and select optimal GGUF/GPTQ/AWQ levels

quantization-claw 4.

FineTune Claw

Plan and execute fine-tuning runs with dataset prep and evaluation

finetune-claw 5.

Dataset Claw

Prepare, clean, augment, and validate datasets for training

dataset-claw 6.

Token Claw AI

Audit tokenizer behavior, design token budgets, and analyze costs

token-claw-ai

How to use this pack

Install the CLI

npm install -g @flickclaw/cli@latest

Install each agent

npm exec --yes @flickclaw/cli@latest -- install ollama-claw --target cursor
npm exec --yes @flickclaw/cli@latest -- install rag-claw --target cursor
npm exec --yes @flickclaw/cli@latest -- install quantization-claw --target cursor
npm exec --yes @flickclaw/cli@latest -- install finetune-claw --target cursor
npm exec --yes @flickclaw/cli@latest -- install dataset-claw --target cursor
npm exec --yes @flickclaw/cli@latest -- install token-claw-ai --target cursor

Run each agent in sequence following the recommended workflow. Each agent produces structured output that feeds into the next.

Browse all agents See Pricing