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

n8n-nodes-webhound

v0.1.2

Published

Run budgeted Webhound reports and datasets, wait for honest completion, and retrieve the result with its evidence.

Readme

n8n-nodes-webhound

This is an n8n community node for Webhound. It starts private reports and datasets with an explicit research budget, waits for truthful completion, and returns either the polished result or the evidence behind it.

Webhound is for questions where the amount of research matters. The dollar budget is part of the task: it controls effort and caps financial exposure. done=true is the authoritative completion signal. output_ready=true by itself may still describe an intermediate artifact.

Hound is a research harness built with DeepSeek V4 Pro and GPT-5.4 across planning, execution, verification, and assembly. It is not a selectable model or a direct pass-through.

Installation

Install n8n-nodes-webhound through Settings → Community Nodes in a supported n8n instance. For local development:

pnpm install
pnpm dev

The development command builds the node, loads it into a local n8n instance, and watches for changes.

Credentials

Every n8n user or workspace supplies its own WEBHOUND_KEY.

  1. Create a Webhound API key at webhound.ai/api.
  2. In n8n, create a Webhound API credential.
  3. Paste the key into WEBHOUND_KEY and run the credential test.

n8n encrypts saved credentials. This node:

  • calls only https://api.webhound.ai/api/v2/mcp;
  • does not accept a configurable endpoint or a publisher-owned key;
  • does not expose n8n's generic Custom API Call proxy;
  • does not read environment variables or files;
  • never writes the key to output or logs;
  • rejects redirects so the credential is not forwarded to another host.

Operations

Start Report

Starts a private cited report. Required inputs:

  • Research Prompt: the question, scope, constraints, and desired artifact;
  • Maximum Budget (USD): an explicit amount from $1 to $500;
  • Confirm Spend: must be set manually to true after the exact budget is approved.

The confirmation is a local guard and is removed before the MCP request. A rejected or unavailable start receives one request only; this node has no automatic retry loop. Spend-bearing starts accept exactly one input item, so one confirmation cannot multiply the approved budget across a batch.

Start Dataset

Starts a private sourced dataset. It has the same budget and confirmation gate as Start Report. An optional JSON object can describe requested fields.

Watch / Wait

Use a zero-second wait for one immediate webhound_watch call. Use 1-110 seconds for one bounded webhound_wait call. Healthy budgeted research can run for a long time. Follow the returned check-in guidance and keep waiting until done=true.

Get Output

Returns the complete report output, a selected report document, or dataset rows. Leave Allow Partial off unless someone explicitly asked for an interim snapshot.

Get Evidence Pack

Returns the final output plus working documents, claim traces, and sources. Use this for diligence, critique, decisions, or other work where the investigation matters as much as the final prose.

Account

Returns credits, recent usage, free-run status, and current account defaults without spend.

Help

Returns current Webhound guidance for budgets, completion, reports, datasets, evidence, billing, setup, and troubleshooting without spend.

Workflow pattern

  1. Use Start Report or Start Dataset with the approved dollar budget.
  2. Store the returned session_id.
  3. Schedule Watch / Wait based on runtime_estimate.recommended_next_check_seconds.
  4. Continue until done is true.
  5. Use Get Output for the finished artifact or Get Evidence Pack for the full research trail.

Avoid a tight polling loop. A normal long-running session is expected to return done=false many times.

Output

The node returns Webhound's structured response as the n8n item JSON and adds mcp_summary, the concise text returned by the MCP tool. Alerts, budget progress, runtime estimates, sources, claims, rows, and documents remain available in their original structured fields.

Compatibility

  • Development and CI: Node.js 22.22.0
  • n8n node API: 1
  • Package style: current @n8n/node-cli
  • Transport: stateless MCP Streamable HTTP, protocol version 2025-06-18

Development

pnpm install --frozen-lockfile
pnpm lint
pnpm typecheck
pnpm test
pnpm audit --audit-level high
pnpm pack --dry-run

No live Webhound key is required by the unit tests. They use mocked n8n request helpers and assert that secrets are redacted, redirects are rejected, spend confirmation is checked before the network, and failed starts are not retried.

Resources

Version history

  • 0.1.2: uses n8n-native operation errors throughout so Creator Portal verification can classify failures correctly.
  • 0.1.1: provenance release via npm trusted publishing; excludes the TypeScript build cache from the package.
  • 0.1.0: initial report, dataset, watch/wait, output, evidence, account, and help actions.