B2B Sales Intelligence Agent
Pricing
from $200.00 / 1,000 lead enricheds
B2B Sales Intelligence Agent
AI-powered B2B lead research agent. Scrapes company websites, scores ICP fit (0–100), surfaces buying signals from news, and generates personalised cold email + LinkedIn outreach copy. Powered by LangGraph + OpenRouter.
Pricing
from $200.00 / 1,000 lead enricheds
Rating
0.0
(0)
Developer
JOSEPH
Maintained by CommunityActor stats
0
Bookmarked
2
Total users
0
Monthly active users
4 days ago
Last modified
Categories
Share
B2B Sales Intelligence & Lead Qualification Agent
Turn a list of company domains into a full sales intelligence dossier — complete with ICP scores, buying signals, and personalised outreach copy — in seconds.
Powered by LangGraph + OpenRouter (supports GPT-4o, GPT-4o-mini, and more).
What it does
For each company you provide, the agent runs a 5-step research pipeline:
- Website scrape — Fetches the homepage and About page to understand what the company actually does
- Firmographic enrichment — Pulls tech stack data (via BuiltWith) and company details
- News signals — Fetches recent headlines from Google News to surface buying triggers (fundraising, hiring, product launches, leadership changes)
- ICP scoring — Scores the company 0–100 on how well it fits your Ideal Customer Profile, with detailed reasoning
- Outreach generation — Writes a personalised cold email (150–200 words) and LinkedIn message (60–80 words) referencing real news signals
All results are pushed to a structured dataset you can export as JSON, CSV, or Excel.
Who is this for
- SDRs and AEs who want to walk into a call knowing the company inside-out
- Sales ops teams automating prospect research at scale
- Lead generation agencies delivering enriched lists to clients
- Founders doing outbound without a sales team
Input
| Field | Type | Required | Description |
|---|---|---|---|
leads | Array | Yes | List of { domain, companyName } objects |
icpCriteria | Object | No | Your ICP definition (industries, sizes, regions, pain points) |
modelName | String | No | LLM model: gpt-4o-mini (default), gpt-4o, o1, o3-mini |
openaiApiKey | String | Yes | Your OpenRouter API key (get one free at openrouter.ai) |
Example input
{"leads": [{ "domain": "stripe.com", "companyName": "Stripe" },{ "domain": "notion.so", "companyName": "Notion" },{ "domain": "linear.app", "companyName": "Linear" }],"icpCriteria": {"target_industries": ["SaaS", "FinTech", "Developer Tools"],"target_company_sizes": ["51-200", "201-500"],"target_regions": ["North America", "Europe"],"pain_points": ["manual workflows", "team collaboration", "scaling engineering"]},"modelName": "gpt-4o-mini","openaiApiKey": "YOUR_OPENROUTER_API_KEY"}
Output
Each lead produces one row in the dataset:
| Field | Type | Description |
|---|---|---|
domain | string | Company domain |
companyName | string | Company name |
icpScore | number | ICP fit score 0–100 |
icpReasoning | string | Why this score was given |
firmographics | object | About page text + tech stack snippet |
newsSignals | array | Recent news headlines |
outreachEmail | string | Ready-to-send cold email |
outreachLinkedIn | string | Ready-to-send LinkedIn message |
error | string | Error message if a lead failed, null otherwise |
Pricing
This actor uses Pay-Per-Event (PPE) pricing:
| Event | Cost |
|---|---|
| Actor start | $0.10 per run |
| Lead enriched | $1.00 per successfully processed lead |
A run with 10 leads costs roughly $10.10 — far cheaper than Apollo, ZoomInfo, or Clay for equivalent personalised research.
How to get your OpenRouter API key
- Go to openrouter.ai
- Sign up for a free account
- Navigate to Keys → Create Key
- Paste the key into the
openaiApiKeyfield
OpenRouter supports GPT-4o, Claude 3.5, Gemini, and many more models on a pay-as-you-go basis.
Notes
- The agent uses Google News RSS and public web pages only — no paid data providers required
- Results quality improves with a more detailed
icpCriteriaobject - For best results use
gpt-4o; for fastest/cheapest usegpt-4o-mini - Each lead takes approximately 15–40 seconds depending on the model and website speed