Babysitter semantic-kernel-setup
Microsoft Semantic Kernel planner and plugin setup for orchestrated AI
install
source · Clone the upstream repo
git clone https://github.com/a5c-ai/babysitter
Claude Code · Install into ~/.claude/skills/
T=$(mktemp -d) && git clone --depth=1 https://github.com/a5c-ai/babysitter "$T" && mkdir -p ~/.claude/skills && cp -r "$T/library/specializations/ai-agents-conversational/skills/semantic-kernel-setup" ~/.claude/skills/a5c-ai-babysitter-semantic-kernel-setup && rm -rf "$T"
manifest:
library/specializations/ai-agents-conversational/skills/semantic-kernel-setup/SKILL.mdsource content
Semantic Kernel Setup Skill
Capabilities
- Configure Semantic Kernel with AI services
- Create semantic and native functions (plugins)
- Set up planners (Handlebars, Stepwise)
- Implement memory connectors
- Design kernel function chains
- Configure automatic function calling
Target Processes
- function-calling-agent
- plan-and-execute-agent
Implementation Details
Core Components
- Kernel: Central orchestrator
- Plugins: Collections of functions
- Planners: Goal to action decomposition
- Memory: Context and semantic storage
Planner Types
- Handlebars Planner
- Stepwise Planner
- Function Calling Stepwise
Configuration Options
- AI service connectors (OpenAI, Azure)
- Plugin registration
- Planner selection
- Memory backend
- Logging and telemetry
Best Practices
- Clear function descriptions
- Appropriate planner selection
- Plugin organization
- Error handling patterns
Dependencies
- semantic-kernel