
Write HTML. Render video. Built for agents.
Zypher Agent is a framework for building production-ready agents in TypeScript. It provides a reactive agent loop, built-in file and terminal tools, MCP support, model provider support, and checkpoint handling.
Builders who want to add agent behavior to their apps and keep control over tools, models, and execution flow.
You can ship agents that act on tools, use MCP servers, and keep checkpoints you can review or revert.
The agent decides the next step from LLM output instead of following a fixed workflow.
You can track, review, and revert agent changes with built-in checkpoint management.
It includes file, search, and terminal tools, and lets you add custom tools.
It can connect to MCP servers and supports OAuth authentication.
It works with Anthropic Claude and OpenAI models through one interface.
You can customize behavior after inference with interceptor hooks.
It includes timeouts, concurrency protection, and error handling for app use.
deno add jsr:@zypher/agent
Production-ready AI agents that live in your applications
[!NOTE] Support for npm coming soon.
deno add jsr:@zypher/agent
import { createZypherAgent } from "@zypher/agent";
import { createFileSystemTools } from "@zypher/agent/tools";
import { eachValueFrom } from "rxjs-for-await";
const agent = await createZypherAgent({
model: "claude-sonnet-4-5-20250929", // Or use ModelProvider for advanced options
tools: [...createFileSystemTools()],
mcpServers: ["@modelcontextprotocol/sequentialthinking-server"],
});
// Run task with streaming
const taskEvents = agent.runTask("Implement authentication middleware");
for await (const event of eachValueFrom(taskEvents)) {
console.log(event);
}
See our documentation for full usage examples and API reference.
Licensed under the Apache License, Version 2.0. See LICENSE.md for details.
Built with ♥️ by CoreSpeed
Sign in to join the discussion.
No comments yet. Be the first to say what this is good for.

Write HTML. Render video. Built for agents.
Ultra-lightweight, open-source, self-hosted personal AI agent framework in Python with WebUI, tools, memory, MCP, multi-agent workflows, automation, and chat apps
SkillOpt is a text-space optimizer that trains reusable natural-language skills for frozen LLM agents through trajectory-driven edits, validation-gated updates, and deployable best_skill.md artifacts.

Omnigent is an open-source AI agent framework and meta-harness: orchestrate Claude Code, Codex, Cursor, Pi, and custom agents — swap harnesses without rewriting, enforce policies and sandboxing, and collaborate in real time from any device.
A theoretical reconstruction of the Claude Mythos architecture, built from first principles using the available research literature.
🕷️ An adaptive Web Scraping framework that handles everything from a single request to a full-scale crawl!