Command Summarizer

---

name: agent-brain-summarizer description: Configure the summarization provider for code summaries parameters:

• name: provider description: Summarization provider (anthropic, openai, gemini, grok, ollama) required: false
• name: model description: Model name for the provider required: false skills:
• using-agent-brain

---

Agent Brain Summarizer Configuration

Purpose

Configures the summarization provider used during document indexing. Summarization generates concise descriptions of code and documents to improve search relevance.

Usage

/agent-brain-summarizer [provider] [--model <model>]

Parameters

Parameter	Required	Default	Description
provider	No	-	Provider: anthropic, openai, gemini, grok, ollama
--model	No	Provider default	Specific model to use

Available Providers

Anthropic

High-quality code-aware summarization.

Model	Speed	Use Case
claude-haiku-4-5-20251001	Fast	Cost-effective, good quality
claude-sonnet-4-5-20250514	Medium	Balanced quality/speed
claude-opus-4-5-20251101	Slower	Highest quality

Configuration:

export SUMMARIZATION_PROVIDER=anthropic
export SUMMARIZATION_MODEL=claude-haiku-4-5-20251001
export ANTHROPIC_API_KEY=sk-ant-...

OpenAI

Versatile summarization with good code understanding.

Model	Speed	Use Case
gpt-5-mini	Fast	Cost-effective
gpt-5	Medium	High quality

Configuration:

export SUMMARIZATION_PROVIDER=openai
export SUMMARIZATION_MODEL=gpt-5-mini
export OPENAI_API_KEY=sk-proj-...

Gemini

Google's models with large context windows.

Model	Speed	Use Case
gemini-3-flash	Fast	Cost-effective
gemini-3-pro	Medium	Higher quality

Configuration:

export SUMMARIZATION_PROVIDER=gemini
export SUMMARIZATION_MODEL=gemini-3-flash
export GOOGLE_API_KEY=...

Grok

xAI's conversational model.

Model	Speed	Use Case
grok-4	Medium	General use

Configuration:

export SUMMARIZATION_PROVIDER=grok
export SUMMARIZATION_MODEL=grok-4
export XAI_API_KEY=...

Ollama (Local)

Privacy-first local summarization.

Model	Speed	Use Case
llama4:scout	Fast	General purpose, lightweight
mistral-small3.2	Fast	Balanced
qwen3-coder	Medium	Code-focused
gemma3	Fast	Efficient
deepseek-coder-v3	Medium	Code-optimized

Configuration:

export SUMMARIZATION_PROVIDER=ollama
export SUMMARIZATION_MODEL=llama4:scout
export OLLAMA_BASE_URL=http://localhost:11434

Setup:

# Pull the model first
ollama pull llama4:scout

Execution

Interactive Configuration

If no provider is specified, use AskUserQuestion:

Which summarization provider would you like to use?

Options:
1. Anthropic (claude-haiku-4-5-20251001) - High quality, code-aware (Recommended)
2. OpenAI (gpt-5-mini) - Fast, cost-effective
3. Gemini (gemini-3-flash) - Large context support
4. Grok (grok-4) - xAI's model
5. Ollama (llama4:scout) - Local, no API key required

Direct Configuration

# Set to Anthropic
/agent-brain-summarizer anthropic --model claude-haiku-4-5-20251001

# Set to OpenAI
/agent-brain-summarizer openai --model gpt-5-mini

# Set to Ollama
/agent-brain-summarizer ollama --model llama4:scout

Apply Configuration

Generate the export commands:

# Add to your shell profile or .env file:
export SUMMARIZATION_PROVIDER=anthropic
export SUMMARIZATION_MODEL=claude-haiku-4-5-20251001

Or update the project configuration:

agent-brain config set summarization_provider anthropic
agent-brain config set summarization_model claude-haiku-4-5-20251001

Post-Configuration

After changing summarization providers, you may want to re-index to use the new summarizer:

# Re-index to apply new summarization
agent-brain reset --yes
agent-brain index /path/to/docs

Note: Changing summarization provider doesn't require re-indexing for existing searches to work, but new summaries won't be generated until re-indexing.

Verification

# Verify summarization provider is working
agent-brain verify

# Test summarization
agent-brain test-summarize "def hello_world():\n    print('Hello, World!')"

Output

Configuration Applied

Summarization Provider Configuration
====================================
Provider: anthropic
Model: claude-haiku-4-5-20251001
API Key: ANTHROPIC_API_KEY (configured)

Configuration saved. Re-index recommended for new summaries.

Provider Comparison

Summarization Provider Comparison
=================================
                  | Anthropic     | OpenAI        | Gemini        | Ollama
------------------|---------------|---------------|---------------|-------------
Quality           | Excellent     | Very Good     | Good          | Good
Code Awareness    | Excellent     | Very Good     | Good          | Varies
Speed             | Fast          | Fast          | Fast          | Varies
Cost (1M tokens)  | $0.80 input   | $0.50 input   | $0.10 input   | Free
Privacy           | Cloud         | Cloud         | Cloud         | Local

Error Handling

Invalid Provider

Error: Unknown provider 'xyz'.
Valid options: anthropic, openai, gemini, grok, ollama

Model Not Available

Error: Model 'invalid-model' not available for provider 'anthropic'.
Available models: claude-haiku-4-5-20251001, claude-sonnet-4-5-20250514, claude-opus-4-5-20251101

Ollama Not Running

Error: Cannot connect to Ollama at http://localhost:11434
Resolution: Start Ollama with 'ollama serve' or check OLLAMA_BASE_URL

Missing API Key

Error: ANTHROPIC_API_KEY not set for Anthropic provider.
Resolution: export ANTHROPIC_API_KEY="sk-ant-..."

Related Commands

• /agent-brain-providers - List all available providers
• /agent-brain-embeddings - Configure embedding provider
• /agent-brain-verify - Verify configuration