Global AI Network

NASA Patent Lead Generator Agent

Automatically discover NASA patents, find qualified commercial partners using Google Search, and generate AI-powered outreach emails with leads saved to Notion.

21+
Total Deployments
15 min
Setup Time
v1.0
Version

Need Help Getting Started? Our AI Specialists Will Set It Up For Free

1-Click Deployment 5-Min Setup Free Expert Support
Technology Partners

Required Integrations

This agent works seamlessly with these platforms to deliver powerful automation.

Browser Automation

Browser Automation

Automate web browsers with AI-powered interactions. Navigate pages, fill forms,...

Google Search

Google Search

Search the web, specific sites, or images using Google Custom Search API. Perfec...

Notion

Notion

All-in-one workspace for content and collaboration with powerful database and pa...

OpenAI

OpenAI

Leverage OpenAI's powerful language models to generate text, answer questions, a...

Step by Step

Setup Tutorial

mission-briefing.md

What This Agent Does

This intelligent workflow automates the process of identifying and qualifying potential commercial partners for NASA patents. It searches through NASA's patent database, finds companies that might benefit from the technology, evaluates their fit using AI analysis, and automatically creates qualified leads in your Notion database with personalized outreach emails.

Key benefits include:

  • Saves 5-10 hours per week on manual patent commercialization research
  • Automatically qualifies leads using AI-powered analysis of company websites and patent descriptions
  • Generates personalized outreach emails tailored to each company's business model
  • Organizes everything in Notion with LinkedIn profiles and fit scores for easy follow-up

Perfect for: NASA technology transfer offices, patent commercialization teams, business development professionals working with government IP, and innovation scouts looking to connect cutting-edge technology with commercial applications.

Who Is It For

This workflow is designed for:

  • Technology Transfer Specialists at NASA or other research institutions who need to identify commercial partners for patented technologies
  • Business Development Teams responsible for licensing NASA patents to private sector companies
  • Innovation Managers seeking to match breakthrough technologies with companies that have the resources to commercialize them
  • Patent Commercialization Consultants who work with government agencies to maximize the commercial impact of publicly-funded research

You'll get the most value from this agent if you regularly need to research potential licensees, struggle with the time-consuming nature of company research, or want to scale your outreach efforts while maintaining personalization.

Required Integrations

Google Search

Why it's needed: This integration powers the discovery of potential partner companies and their LinkedIn profiles. It searches the web for companies relevant to each patent's technology area and finds decision-maker profiles for outreach.

Setup steps:

  1. Navigate to the Integrations page in TaskAGI
  2. Click Connect next to Google Search
  3. Choose between using the built-in Google Search API or a custom search engine
  4. For custom search (recommended for higher limits):
  5. Paste both values into TaskAGI's integration settings
  6. Click Save & Test Connection

Configuration tip: The free tier provides 100 searches per day. For production use with multiple patents, consider upgrading to a paid plan.

Browser Automation

Why it's needed: This integration crawls company websites to extract detailed information about their business model, products, and strategic focus. This data feeds into the AI analysis that determines partnership fit.

Setup steps:

  1. Go to Integrations in TaskAGI
  2. Find Browser Automation and click Connect
  3. TaskAGI's built-in browser automation requires no external API keys
  4. Simply toggle the integration to Enabled
  5. Configure timeout settings (recommended: 30 seconds)
  6. Enable JavaScript rendering for modern websites

Note: This integration runs on TaskAGI's infrastructure, so no additional setup is required. It automatically handles dynamic content and JavaScript-heavy sites.

OpenAI

Why it's needed: The GPT-4 model analyzes the patent description alongside company information to determine fit score, identify specific use cases, and draft personalized outreach emails. This is the "intelligence" layer that makes qualification decisions.

Setup steps:

  1. Visit OpenAI's platform
  2. Create an account or sign in
  3. Navigate to API Keys in your account settings
  4. Click Create new secret key
  5. Name it TaskAGI-NASA-Patents for easy identification
  6. Copy the key immediately (it won't be shown again)
  7. In TaskAGI, go to IntegrationsOpenAI
  8. Paste your API key
  9. Select GPT-4o as your default model (required for this workflow)
  10. Click Save & Test Connection

Cost consideration: GPT-4o costs approximately $0.01-0.03 per patent analyzed. Budget accordingly based on your expected volume.

Notion

Why it's needed: Notion serves as your centralized lead database where all qualified companies are stored with their fit scores, draft emails, LinkedIn profiles, and patent information for easy team collaboration and follow-up.

Setup steps:

  1. Log into your Notion workspace
  2. Create a new database called NASA Patent Leads (or use an existing one)
  3. Add these properties to your database:
    • Company Name (Title)
    • Patent (Text)
    • Fit Score (Number)
    • Draft Email (Text - Long)
    • Company URL (URL)
    • LinkedIn Profile (URL)
    • Date Added (Date)
  4. Visit Notion's integrations page
  5. Click Create new integration
  6. Name it TaskAGI Integration
  7. Select your workspace
  8. Copy the Internal Integration Token
  9. In your Notion database, click ShareInvite → Select your integration
  10. In TaskAGI, go to IntegrationsNotion
  11. Paste your integration token
  12. Click Authorize and select your database
  13. Map the database properties to match the workflow fields

Configuration Steps

Step 1: Configure the Manual Trigger

The workflow starts with a manual trigger, allowing you to run it on-demand. No configuration needed here—simply use this to initiate the workflow whenever you want to process patents.

Step 2: Set Up Configuration Parameters

In the Configuration node, define these key parameters:

  • patent_category: The NASA technology category to search (e.g., "aerospace", "materials", "robotics")
  • max_patents: Number of patents to process per run (recommended: 5-10 for testing, 20-50 for production)
  • min_fit_score: Minimum score threshold for creating Notion leads (recommended: 7 on a 1-10 scale)

Example configuration:

{
  "patent_category": "aerospace propulsion",
  "max_patents": 10,
  "min_fit_score": 7
}

Step 3: NASA Patents API Request

The NASA Patents API node is pre-configured to query NASA's patent database. Verify these settings:

  • Method: GET
  • URL: https://api.nasa.gov/techtransfer/patent/
  • Query Parameters: Include your NASA API key (get one free at api.nasa.gov)
  • Response Path: results (extracts the patent array)

Add your NASA API key to the headers: api_key: YOUR_NASA_API_KEY

Step 4: Loop Configuration

The Loop Over Patents node iterates through each patent. Set:

  • Loop Array: {{4404.output.results}} (references the API response)
  • Item Variable Name: patent
  • Max Iterations: Match your max_patents value

Step 5: Company Search Settings

In Google Search - Find Company, configure:

  • Search Query: {{patent.title}} company commercial application
  • Number of Results: 5
  • Safe Search: off

This finds companies working in the patent's technology area.

Step 6: URL Extraction Function

The Extract Company URL node uses JavaScript to parse search results:

const results = input.searchResults;
return results[0]?.url || null;

This extracts the top company website URL. No changes needed unless you want to modify selection logic.

Step 7: Conditional URL Check

Check URL Found verifies a company was discovered:

  • Condition: {{4407.output}} !== null
  • True Path: Continue to website crawling
  • False Path: Skip to No URL Found node

Step 8: Website Crawling Configuration

Crawl Company Website settings:

  • URL: {{4407.output}}
  • Wait for Load: 3000 ms
  • Extract: text (full page content)
  • Max Length: 5000 characters (prevents token overload)

Step 9: AI Analysis Setup

The Analyze Fit & Draft Email node is the workflow's intelligence center. Verify:

  • Model: gpt-4o (required for quality analysis)
  • Temperature: 0.7 (balanced creativity)
  • Max Tokens: 1000
  • Prompt: The pre-configured prompt analyzes fit and drafts emails. Key variables:
    • {{patent.title}}
    • {{patent.description}}
    • {{4408.output.content}}

The AI returns JSON with fitScore, reasoning, and draftEmail.

Step 10: Score Filtering

High Score Filter checks if the lead qualifies:

  • Condition: {{4409.output.fitScore}} >= {{config.min_fit_score}}
  • True Path: Create Notion entry
  • False Path: Skip lead

Step 11: LinkedIn Profile Search

Search LinkedIn Profile finds decision-makers:

  • Site: linkedin.com
  • Query: {{patent.title}} CEO OR "Chief Technology Officer" OR "VP Innovation"
  • Results: 1

Step 12: Notion Data Preparation

Prepare Notion Data formats the information:

return {
  companyName: input.companyUrl.split('/')[2],
  patent: input.patent.title,
  fitScore: input.analysis.fitScore,
  draftEmail: input.analysis.draftEmail,
  companyUrl: input.companyUrl,
  linkedinProfile: input.linkedin[0]?.url || "Not found"
};

Step 13: Create Notion Lead

Create Notion Lead maps the data to your database properties. Ensure field mappings match your Notion database structure exactly.

Testing Your Agent

Initial Test Run

  1. Click Test Workflow in the top-right corner
  2. Start with max_patents: 2 to limit API costs
  3. Monitor the execution panel as each node processes
  4. Check for green checkmarks indicating successful completion

Verification Checklist

After NASA API call:

  • ✅ Verify patents are returned (check node output)
  • ✅ Confirm patent descriptions are complete

After Google Search:

  • ✅ Review that relevant companies appear in results
  • ✅ Check that URLs are being extracted

After Website Crawl:

  • ✅ Ensure website content is captured (not empty)
  • ✅ Verify content length is reasonable (not truncated mid-sentence)

After AI Analysis:

  • ✅ Review fit scores for reasonableness (should vary between patents)
  • ✅ Read draft emails to ensure they're personalized and relevant
  • ✅ Check that reasoning explains the score

After Notion Creation:

  • ✅ Open your Notion database
  • ✅ Verify all fields are populated correctly
  • ✅ Confirm LinkedIn profiles link to relevant people

Expected Results

A successful run should:

  • Process all specified patents within 3-5 minutes
  • Create Notion entries for 30-50% of patents (depending on your threshold)
  • Generate unique, contextual emails for each qualified lead
  • Include working URLs and LinkedIn profiles

Troubleshooting

"NASA API returned no results"

Cause: Invalid API key or overly specific search category

Solution:

  • Verify your NASA API key at api.nasa.gov
  • Broaden your patent_category (try "aerospace" instead of "hypersonic propulsion systems")
  • Check NASA's API status page for outages

"Google Search quota exceeded"

Cause: You've hit the daily search limit (100 for free tier)

Solution:

  • Reduce max_patents to process fewer patents per run
  • Upgrade to Google Custom Search paid tier ($5 per 1,000 queries)
  • Schedule workflow runs to spread searches across multiple days

"OpenAI timeout or error"

Cause: Prompt too long, model unavailable, or insufficient credits

Solution:

  • Reduce max_length in the website crawl node to 3000 characters
  • Check your OpenAI account has available credits
  • Verify you have access to GPT-4o (may require paid account)
  • Try reducing max_tokens to 800

"Notion entry creation failed"

Cause: Field mapping mismatch or permission issues

Solution:

  • Verify your Notion integration has write access to the database
  • Check that all database properties exist and match names exactly (case-sensitive)
  • Ensure property types match (Text, Number, URL, etc.)
  • Re-authorize the Notion integration if recently changed

"Fit scores are all very high or very low"

Cause: AI prompt needs calibration for your specific needs

Solution:

  • Adjust the prompt to emphasize your specific criteria
  • Modify the scoring rubric in the AI prompt
  • Change min_fit_score threshold to match actual score distribution
  • Add more specific industry or company size requirements to the prompt

"LinkedIn profiles are not found"

Cause: Search query too specific or LinkedIn blocking automated searches

Solution:

  • Broaden the LinkedIn search query to include more job titles
  • Add a delay between searches (use a Wait node before LinkedIn search)
  • Consider using LinkedIn's official API for more reliable results
  • Manually add profiles as a follow-up step for high-value leads

Next Steps

After Successful Setup

  1. Run a production batch: Increase max_patents to 20-50 and process a full category
  2. Review lead quality: Analyze your Notion database to see which patents generated the best matches
  3. Refine scoring criteria: Adjust the AI prompt based on which leads your team actually pursues
  4. Set up scheduling: Use TaskAGI's scheduler to run this workflow weekly or monthly automatically

Optimization Suggestions

Improve lead quality:

  • Add company size filters (target Fortune 500 or startups specifically)
  • Include industry-specific keywords in the Google search query
  • Enhance the AI prompt with your organization's specific partnership criteria

Scale your operations:

  • Create separate workflows for different patent categories
  • Add a Google Sheets integration to track outreach responses
  • Implement email automation to send draft emails directly (with human approval)

Enhance data collection:

  • Add a node to capture company funding information from Crunchbase
  • Include patent citation analysis to prioritize high-impact technologies
  • Scrape company press releases for recent innovation initiatives

Advanced Usage Tips

Multi-stage qualification: Add a second AI analysis node that compares multiple companies for the same patent and ranks them.

Team collaboration: Set up Notion automations that notify team members when high-scoring leads (9+) are added.

Performance tracking: Create a dashboard in Notion that shows conversion rates from lead to partnership by patent category.

Custom scoring models: Replace the AI analysis with a custom scoring algorithm based on your historical partnership data.

Integration with CRM: Add a final node that pushes qualified leads to Salesforce or HubSpot for formal sales tracking.

This workflow transforms patent commercialization from a manual, time-intensive process into an automated, scalable system. Start small, refine based on results, and gradually expand to cover more patent categories as you build confidence in the system. Happy automating! 🚀