Pre-tuned agent personas that work out of the box — skills are what your agent IS, tools are what it CAN DO.
researcher, coder, planner, critic, and summarizer are battle-tested behavioral profiles; activate one and your agent immediately behaves like a specialistcomposeSkills to merge prompts, tools, and delegates; build a research-and-code pipeline in one lineplanner skill knows how to coordinate researcher and coder as sub-agents, so you get multi-agent workflows without writing orchestration codenpm install @agentskit/skills
import { createRuntime } from '@agentskit/runtime'
import { anthropic } from '@agentskit/adapters'
import { researcher, coder, composeSkills } from '@agentskit/skills'
import { webSearch, filesystem } from '@agentskit/tools'
const runtime = createRuntime({
adapter: anthropic({ apiKey: process.env.ANTHROPIC_API_KEY, model: 'claude-sonnet-4-6' }),
tools: [webSearch(), ...filesystem({ basePath: './workspace' })],
})
const result = await runtime.run('Research best practices for TypeScript error handling and write an example', {
skill: composeSkills(researcher, coder),
})
console.log(result.content)
composeSkills to merge multiple personas; add sandbox tools from @agentskit/sandbox for code execution when the skill needs a VM@agentskit/templates createSkillTemplate| Package | Role |
|---|---|
| @agentskit/runtime | createRuntime, skill option |
| @agentskit/tools | Tools skills orchestrate |
| @agentskit/adapters | LLM backends |
| @agentskit/templates | createSkillTemplate, scaffold |