typesafe-ai/skills

Agent skills for building with TypeSafe's System One API

What it solves

It provides a set of agent skills that allow AI agents to integrate with TypeSafe, enabling them to make typed decisions and handle probabilities from "System One" models.

How it works

These skills are installed as plugins for agents (such as Claude Code) or via a general skills installer (skills.sh). Once installed, the agent can use the the typesafe-ai skill to design workflows, access documentation, and compose typed judgments directly in code.

Who it’s for

Developers building with TypeSafe who want to empower their AI agents to handle structured, typed decision-making processes.

Highlights

  • Support for Claude Code as a plugin.
  • Integration via skills.sh for other agents.
  • Ability to route tasks (e.g., support tickets) based on typed decisions with human-in-the-loop review for uncertainty.

Written about in

Related

  • Project
    Zleap-AI/SAG

    A knowledge base application using a novel SAG architecture that combines semantic retrieval and relational reasoning to outperform traditional RAG and GraphRAG.

  • Project
    sdyckjq-lab/llm-wiki-skill

    A system for building a structured, interconnected knowledge base from fragmented data, featuring an interactive knowledge graph and integration with AI CLI agents.

  • Project
    neavo/LinguaGacha

    An AI-powered translator for novels, games, and subtitles that uses an agentic workflow to ensure terminology consistency and formatting retention.

  • Project
    langchain-ai/langchain-nextjs-template

    A starter template for building AI applications with LangChain.js and Next.js, featuring implementations for streaming chat, RAG, and agentic workflows.