Агенты и мультиагенты

TypeSafe Agent Skills

typesafe-ai/skills

Official agent skills for designing applications around TypeSafe typed decisions. Development guidance complements the runtime SDKs and API.

★ 2KЗвёзды
⑂ 109Форки
12Открытые проблемы
Язык
MITЛицензия
Q84Редакционная оценка

Обзор

Official agent skills for designing applications around TypeSafe typed decisions. Development guidance complements the runtime SDKs and API. Start with a bounded task, such as choosing a destination for a support ticket, and let application logic control the next action.

Основные функции

  • Official repository
  • Typed decision workflows
  • Documentation and examples
  • Decisions composed in application code
  • Works with Jev
  • MIT-licensed code

Требования, установка и быстрый старт

npx skills add typesafe-ai/skills --skill typesafe-ai. Review the repository README and dependencies before installation.

Использование

After installation, ask your agent to consult the material when designing a TypeSafe workflow. Runtime API calls still require credentials and a billing review.

Совместимость моделей и варианты использования

For the TypeSafe System One API; this is not Jev weights or a local inference engine.

Лицензия и примечания о рисках

The repository uses the MIT license. API access, billing and service terms apply separately.

Official client/development repository; not model weights. API access and charges apply separately.

Firecrawl

firecrawl/firecrawl

★ 161,1KTypeScript

LangChain

langchain-ai/langchain

★ 143,6KPython

RAGFlow

infiniflow/ragflow

★ 86,7KGo