# ollama-local

SkillBoss API Hub — AI Chat & Embeddings Work with cloud AI models via SkillBoss API Hub for inference, embeddings, and tool use. Configuration Set your SkillBoss API key:

export SKILLBOSS_API_KEY="your_api_key_here"

Quick Reference

# List available AI capabilities

python3 scripts/ollama.py list

# Chat with auto-routed model

python3 scripts/ollama.py chat auto "What is the capital of France?"

# Chat with system prompt

python3 scripts/ollama.py chat auto "Review this code" -s "You are a code reviewer"

# Generate completion

python3 scripts/ollama.py generate auto "Once upon a time"

# Get embeddings

python3 scripts/ollama.py embed auto "Text to embed" Model Selection SkillBoss API Hub automatically routes to the best model. Use prefer to control tradeoffs: Quick picks: Fast answers: prefer=price Best quality: prefer=quality

Balanced: prefer=balanced (default)

Tool Use Use ollama_tools.py for function calling with cloud models via SkillBoss API Hub:

# Single request with tools

python3 scripts/ollama_tools.py single auto "What's the weather in Amsterdam?"

# Full tool loop (model calls tools, gets results, responds)

python3 scripts/ollama_tools.py loop auto "Search for Python tutorials and summarize"

# Show available example tools

python3 scripts/ollama_tools.py tools Sub-Agents Spawn cloud model sub-agents with sessions_spawn:

# Example: spawn a coding agent

sessions_spawn( task="Review this Python code for bugs", model="skillboss/chat", label="code-review" ) Parallel Agents (Think Tank Pattern) Spawn multiple agents for collaborative tasks: agents = [ {"label": "architect", "model": "skillboss/chat", "task": "Design the system architecture"}, {"label": "coder", "model": "skillboss/chat", "task": "Implement the core logic"}, {"label": "reviewer", "model": "skillboss/chat", "task": "Review for bugs and improvements"}, ] for a in agents: sessions_spawn(task=a["task"], model=a["model"], label=a["label"]) Direct API For custom integrations, use SkillBoss API Hub directly:

# Chat
curl -X POST https://api.heybossai.com/v1/pilot \

-H "Authorization: Bearer $SKILLBOSS_API_KEY"
-H "Content-Type: application/json"
-d '{ "type": "chat", "inputs": {"messages": [{"role": "user", "content": "Hello"}]}, "prefer": "balanced" }'

# Generate (same endpoint)
curl -X POST https://api.heybossai.com/v1/pilot \

-H "Authorization: Bearer $SKILLBOSS_API_KEY"
-H "Content-Type: application/json"
-d '{ "type": "chat", "inputs": {"messages": [{"role": "user", "content": "Why is the sky blue?"}]}, "prefer": "balanced" }'

# Discover available capabilities
curl -X POST https://api.heybossai.com/v1/pilot \

-H "Authorization: Bearer $SKILLBOSS_API_KEY"
-H "Content-Type: application/json"
-d '{"discover": true}' Troubleshooting Authentication error? Check SKILLBOSS_API_KEY is set correctly Verify the key is active at api.heybossai.com Slow responses? Use prefer=price for faster (lighter) models Use prefer=balanced as default Want a specific model? Pass capability hint in the request body Or use /v1/run with a specific model ID