Skills

Configure LLM-backed micro-agent capabilities with system prompts, schemas, and execution tiers.

Skills are companion agent capabilities - LLM-backed micro-agents that execute specific tasks within a conversation. Unlike Classic API tools (which run versioned code packages), skills are configured declaratively with a system prompt, input/output schema, and model selection.

Different from Classic API Tools - Classic API Tools are versioned code packages from Git repos. Platform API Skills are LLM-backed declarative micro-agents with prompt-based configuration. Choose Tools for custom code execution, Skills for LLM-native reasoning.

Key Concepts

• Slug: URL-friendly identifier for the skill (lowercase alphanumeric with hyphens/underscores)

• System Prompt: Instructions for the skill's LLM execution

• Input/Result Schema: JSON Schemas defining the expected inputs and structured outputs

• Version: Auto-incremented on each update - skills are versioned implicitly

Execution Tiers

Skills support four execution modes:

Tier	Description
direct	Single LLM call, no tool use - fastest execution
orchestrated	Multi-turn with tool use, managed by the platform
autonomous	Extended agent loop with checkpointing
browser	Browser-based execution with allowed domains

Integration with Tools

Skills can reference Integration endpoints via the integration_tools field, giving the skill's LLM access to external APIs during execution. They can also define inline static_tools.

Testing

Use the test endpoint to execute a skill in isolation - no conversation required. Useful for validating prompt engineering and schema design before wiring into a context graph.

References

Before modifying or deleting a skill, use the references endpoint to see which context graphs (HSMs) and services depend on it.

API Reference

• Skills