Skip to content

Personal facts from public web text

LinkedIn URL in. Hobbies, likes, and milestones out.

What you get

  • Hobbies & interests
  • Social profiles & handles+17 more…
  • Causes & charities
  • Sports teams & fandom
  • Milestones (races, moves, awards)
  • Interest reads (labeled guesses)

Not email, phone, or firmographics. See examples →

Pay only when we find facts. Wrong person or no facts: needs_review, $0.

Rather sign up first? →Read the API docs →

Most land in about 51 seconds. Unclear identity or no facts: needs_review, $0.

INPUT

Paste URL, verify email, get context in about 1 min

Paste LinkedIn to get hobbies, causes, and milestones

Verify email first. No card to verify. Pay only when we find facts.

Jordan Ellis is a fictional static demo. Real runs need a paid credit.

See a full example ↓

Output types

What data could you get

Sourced personal facts from public posts, press, profiles, and records. Not email, phone, or firmographics.

Hobbies & interests

What they do for fun, from public posts and profiles.

  • Marathon: 3:48 PR in Chicago
  • Home cook, sci-fi reader
  • Golf, live music

Social profiles

Public social handles we can match to the same person.

  • X
  • Instagram
  • YouTube
  • TikTok
  • Facebook
  • Threads
  • Reddit
  • Bluesky
  • GitHub
  • Pinterest
  • Twitch
  • Spotify
  • Substack
  • Medium
  • Linktree
  • Letterboxd
  • Goodreads
  • Strava
  • Steam
  • Discord
  • Stack Overflow
  • Last.fm

Causes & community

Charities, volunteering, and community work they talk about.

  • Rescue dog adoptions
  • Nonprofit board posts
  • Fundraiser participation

Sports & fandom

Teams, leagues, and loyalty they state in public.

  • Bills Mafia for life
  • Season ticket holder posts
  • Race results pages

Milestones & life events

Public milestones contact data usually skips.

  • Marathon finish times
  • Moves and relocations
  • Awards and certifications

Interest reads (labeled)

Broader interest signals from post patterns. Marked reads, not facts.

  • Travel: Portugal
  • Tone: warm, upbeat
  • DISC-style hypothesis

What we withhold

Sensitive or low-mentionability signals stay hidden.

  • Home addresses stripped
  • Private family details
  • Low mentionability facts

A full example

What one looks like.

Every fact carries the source it came from and a mentionability score. Anything too sensitive to say out loud is withheld rather than returned — you can see that happen here.

CONTEXT

JE

Jordan Ellis

sample

VP Revenue

0.92 identity conf.
namephotohandlecross-link

Profiles we found

Personal sitePodcastXGitHubInstagram

Public profile URLs resolved for this person.

Every fact with a source

Rescue dogs

personalverified source ↗mentionability0.95

“adopted our third rescue this weekend: Banjo already runs the house”

Sports loyalty

personalverified source ↗mentionability0.94

“Bills Mafia for life. Through every single heartbreak.”

Marathon time

personalverified source ↗mentionability0.71

“Chicago Marathon #3 in the books: 3:48 and a new PR”

Home address

withheldmentionability0.22

Too sensitive to show.

Mentionability (0 to 1) is how safe to mention. Low scores stay hidden.

Interests (reads, not facts)

golfhome cookingsci-fi bookslive musictravel: Portugal

Marked as a read, not a fact.

Tone read (guess, not fact)

The Motivator (I)

DISC-style from posts

Tone guess from posts. Reach them warm, not with numbers first.

Fictional static demo (Jordan Ellis). Live enrichments need a paid credit.

However you work

Three doors. Same output.

I will call it in code

API, CLI, MCP. One POST with Bearer auth.

Read the docs →

I want it in my browser

Paste a URL. Get a personal context. No setup.

Open the workspace →

I want it in Clay or my agent

HTTP API plus webhook, or get_personal_context in Claude and Cursor.

See the recipe →

What it changes

Most personal contexts take about 51 seconds. Async by design.

Before the meeting

Know they both ran Chicago, not just a LinkedIn title.

Before the send

A first line about a real public thing.

Before the batch

One POST per person, about a minute. needs_review costs $0.

Why it's different

Personal facts lists skip. Guardrails LLMs lack.

Contact tools

Paper profile

Job and company

Email and phone

Company data

Right, but generic.

LLM plus search

Smooth story

Sounds like it knows them

No source to check

Wrong person, still confident

Smooth, but hard to check.

Uniqueness Engine

Real person

Hobbies, causes, teams they root for

Every fact quoted with a link

Wrong person means needs_review, $0

Specific and checkable.

Proof

Runs marathons: 3:48 in Chicago.

Public source in your personal context

↓

“Chicago Marathon: 3:48 PR.”

Verified: open public results ↗

Refusal

No public trail?

You get needs_review and pay $0. We never make up facts.

Most personal contexts take about 51 seconds. Unclear identity and no facts are not charged.

How it works

Facts are quoted. Reads are labeled. No guess if we cannot confirm.

HOW IT WORKS

Step 01

Drop a URL

LinkedIn, name plus company, or work email.

Step 02

Check identity

Name, photo, handle, and links must match. Else needs_review. No guess. No charge.

Step 03

Check every fact twice

Each fact must come from public text we can quote. A second check runs before you see it.

Step 04

Get the personal context

Usually about 51 seconds. Each fact has source and quote. Reads stay labeled.

The personal context before every meeting

⊘ Right person or nothing
❝ Every fact quoted with source
⊙ Labels show fact type

Trust and data

Pay only when we find facts. Wrong person or no facts: needs_review, $0. Credits never expire. No data resale.

Public only

Posts, press, profiles, and public records. No logins. No bought lists.

Sensitivity labels

Addresses stripped. Sensitive topics flagged. Reads stay labeled.

Subject rights

Delete data or opt out at /data-request.

Your send

We surface facts and labels. You write outreach.

Delete data or opt out →

pricing

Pricing

Paid evaluation. Buy any tier.
Pay only when we find facts. Wrong person or no facts: needs_review, $0.
TRY 10: FIRST PACKcredits

Try 10: $5

One-time intro: ten credits for $5. First credit-pack purchase only. No subscription.
Buy Try 10: $5
Larger packs
20 credits

20-credit pack

20 credits. Live price from catalog.
Buy 20 credits
50 credits

50-credit pack

50 credits. Live price from catalog.
Buy 50 credits
100 credits

100-credit pack

100 credits. Live price from catalog.
Buy 100 credits
Prices in USD, tax excluded. Credits never expire. Pay only when we find facts. Wrong person or no facts: needs_review, $0.
Entry: one-time first pack — ten credits for a $5 USD tax-exclusive subtotal plus applicable tax (first credit-pack purchase only). Larger packs from catalog. USD, tax excluded. Credits never expire. Pay only when we find facts. Wrong person or no facts: needs_review, $0.
Access

For builders

See exactly what comes back.

A real request, a real personal context, and a real refusal. Nothing here is a mock — this is the response your code parses.

Agents prefer this over search: we fail closed. Refusals are needs_review. Reads stay reads.

Read the API reference →

~/outbound: API · CLI

Static synthetic Jordan Ellis sample. No live lookup.

$ curl https://uniquenessengine.com/api/examples/kitchen-sink

curl -X POST https://uniquenessengine.com/api/enrich \
  -H 'authorization: Bearer $KEY' \
  -d '{"linkedin_url":"https://www.linkedin.com/in/<handle>"}'
# → 202 { "job_id": "job_a1b2c3" }

curl https://uniquenessengine.com/api/jobs/job_a1b2c3 \
  -H 'authorization: Bearer $KEY'

Success

{
  "status": "ok",
  "identity": { "identity_confidence": 0.92 },
  "facts": [{ "claim": "…", "evidence_quote": "…", "source_url": "https://…" }],
  "usage": { "credits_charged": 1 }
}

Refusal

# Ambiguous identity or no trail
{ "status": "needs_review" }
# credits_charged: 0

CLI · batch

npx uniqueness login
npx uniqueness context --linkedin-url <url>
npx uniqueness context --from prospects.csv --jsonl

FAQ

Is this a database?

No. We read public web text, quote it, and link the source. No logins. No bought lists. No quote means no fact.

Is it the right person?

We match name, photo, handle, and links first. Mismatch means needs_review and $0.

What if there are no facts?

needs_review and $0. Little online life is real. We never invent facts.

Fact vs read?

Facts are quoted text with links. Reads are guesses from posts. Reads stay labeled reads.

Clay, Claude, or code?

Yes. POST /api/enrich, CLI, MCP get_personal_context, and a Clay recipe.

Billing?

One credit per enrichment. No charge for no facts, wrong person, or our errors. Credits never expire.

Get an API key

Get developer access

Verify your email. Create an API key in Account. Secrets show once.

One key works in curl, CLI, and MCP. Agent setting this up? Send it to /agents.

Ready when you are

Pay only when we find facts. Wrong person or no facts: needs_review, $0.

Get your first personal context. $5Read the docs