NoodleSpy Documentation

NoodleSpy is an API for structured online research. It discovers conversations across the web, reads and analyzes them, and returns structured findings — themes, direct quotes, and source links — ready for your AI workflow.

What NoodleSpy does

  • Research Runs — Deep async research on any topic. Submit a topic, get back structured findings with themes, quotes, and sources.
  • Web Search — Google Custom Search via API. Discover URLs across the web or target specific sites.
  • Grounded Search — AI-generated answers grounded in web sources, with citations.
  • Fetch — Retrieve any web page as clean markdown, HTML, or a screenshot.

Who it's for

NoodleSpy is built for developers and AI agents. There are no dashboards or manual workflows. You make API calls, you get structured data back.

Common use cases:

  • AI agents that need to research topics before making decisions
  • Product teams monitoring user sentiment across communities
  • Market research pipelines that run on a schedule
  • Content tools that need grounded, cited information

How it works

  1. Authenticate — Create an API key in your dashboard at /app/api-keys
  2. Search — Use POST /search/web to discover relevant pages
  3. Research — Use POST /runs to kick off deep research on a topic
  4. Poll — Use GET /runs/{id} to check status and retrieve findings
  5. Use the data — Findings come back as structured JSON with themes, quotes, and source links

Base URL

All API requests use:

https://api.noodlespy.com

Authentication

Every request requires an API key passed in the Authorization header:

Authorization: Bearer nspy_your_api_key_here

API keys are created and managed in your dashboard.

Rate limits and quotas

Your subscription includes:

  • 30 research credits/month — Credits scale with depth: quick = 1, standard = 2, deep = 3
  • 1,000 utility API calls/month — Web search, grounded search, and fetch

When you exceed your quota, you'll receive a 429 response. Usage resets at the start of your billing cycle.

Every response includes remaining quota in headers:

X-Usage-Remaining-Credits: 28
X-Usage-Remaining-Utility: 994

Next steps