npm package discovery and stats viewer.

Discover Tips

  • General search

    [free text search, go nuts!]

  • Package details

    pkg:[package-name]

  • User packages

    @[username]

Sponsor

Optimize Toolset

I’ve always been into building performant and accessible sites, but lately I’ve been taking it extremely seriously. So much so that I’ve been building a tool to help me optimize and monitor the sites that I build to make sure that I’m making an attempt to offer the best experience to those who visit them. If you’re into performant, accessible and SEO friendly sites, you might like it too! You can check it out at Optimize Toolset.

About

Hi, 👋, I’m Ryan Hefner  and I built this site for me, and you! The goal of this site was to provide an easy way for me to check the stats on my npm packages, both for prioritizing issues and updates, and to give me a little kick in the pants to keep up on stuff.

As I was building it, I realized that I was actually using the tool to build the tool, and figured I might as well put this out there and hopefully others will find it to be a fast and useful way to search and browse npm packages as I have.

If you’re interested in other things I’m working on, follow me on Twitter or check out the open source projects I’ve been publishing on GitHub.

I am also working on a Twitter bot for this site to tweet the most popular, newest, random packages from npm. Please follow that account now and it will start sending out packages soon–ish.

Open Software & Tools

This site wouldn’t be possible without the immense generosity and tireless efforts from the people who make contributions to the world and share their work via open source initiatives. Thank you 🙏

© 2026 – Pkg Stats / Ryan Hefner

cracked-ai-vercel

v0.1.0

Published

Vercel AI SDK tool definitions for Cracked, the agent tool router: crackedTools() and crackedToolsFor(query).

Readme

Cracked for the Vercel AI SDK

crackedTools() returns AI SDK tool definitions (discover, inspect, run, run_capability) so generateText / streamText agents can reach 50,000+ tools through one API key. crackedToolsFor(query) turns discover results into one tool per endpoint with the provider's real input schema.

Publishing to npm soon. Until then install from the repo path:

cd sdk/typescript && npx tsc && cd ../integrations/vercel-ai && npx tsc   # build both
npm install ./sdk/integrations/vercel-ai ai zod                           # from your app
# later: npm install cracked-ai-vercel ai zod

Set CRACKED_API_KEY (https://cracked.ai/app/keys, or Cracked.registerAgent()).

Use

import { generateText, stepCountIs } from "ai";
import { openai } from "@ai-sdk/openai";
import { crackedTools } from "cracked-ai-vercel";

const { text } = await generateText({
  model: openai("gpt-4.1-mini"),
  tools: crackedTools(),                 // cracked_discover, cracked_inspect, cracked_run, cracked_run_capability
  stopWhen: stepCountIs(6),              // AI SDK v5; use maxSteps: 6 on v4
  system: "Find a tool with cracked_discover, read its schema with cracked_inspect, then cracked_run it. Quote output fields; never invent data.",
  prompt: "What is the weather in Austin, TX right now?",
});

One tool per endpoint, schema from inspect:

import { crackedToolsFor } from "cracked-ai-vercel";
const tools = await crackedToolsFor("instagram profile followers", { limit: 3 });
await generateText({ model, tools, prompt: "How many followers does @nike have?", stopWhen: stepCountIs(3) });

Options: crackedTools({ apiKey, baseUrl, client, maxOutputChars, timeoutMs }).

Tool definitions are plain { description, inputSchema, execute } objects (AI SDK v5 shape) built with zod, so this package does not import ai itself. On AI SDK v4 wrap them: toV4(crackedTools()) maps inputSchema to parameters.

Each tool returns a status line (runId, status, httpStatus, costUsd) then the JSON output, truncated to maxOutputChars (default 20,000). API errors come back as JSON strings (402 = top up; BLOCKED = the provider needs your own key, pick another).

npx tsx example.ts exercises the tools directly (no LLM needed).