Local-only business research
Research a business from many sources. Review every claim.
Give it a name and a domain. One pipeline run reads the business’s own website, public directories, local news and the sources you hold keys for, then turns what it finds into claims you accept, edit or reject. The result is a set of files you commit, one folder per business.
- websiteok
- google_placesnot_configured
- gdeltok
- serperok
- bbbnot_found
- dnbnot_configured
- llm_curateok
data.json · claims.json · meta.json · profile.md written
What it does
- 01
Many sources, one run
The business’s own website, Google Places, GDELT news, Serper directory and social-handle discovery, BBB, D&B and an LLM curation pass, in one pipeline run.
- 02
Claims, not guesses
Every value arrives as a claim with its source, capture method, time and confidence. Disagreeing claims stay side by side until you decide.
- 03
Your decision always wins
A curation decision is never overwritten by a later pipeline run. A new, disagreeing claim is shown to you as a conflict instead.
- 04
Missing keys are not errors
Every API key is optional. A source without its key reports
not_configuredand the run carries on with the rest. - 05
Files, not a database
Each business is a folder of deterministic JSON and Markdown files that you commit to git, so every change is a reviewable diff.
- 06
Runs on your machine
No hosting, no account and no sign-in. The tool runs on localhost and calls each source directly with your own keys.
From a name to a profile
- Step 1
Create a business
Enter a name and a domain. The tool gives the business a bizId and a slug, and adds both to the business index (
data/index.json). - Step 2
Start a pipeline run
The run takes a lock on the business, then attempts each source in turn: the website, BBB, social discovery, Google Places, GDELT, Serper, D&B and NewsAPI. Each one records a source status.
- Step 3
Capture and extract claims
Each connector returns claims. A claim names its field path, value, source, capture method, time and confidence. A blocked page stops there; a person can capture it by hand later.
- Step 4
Merge and resolve
Fresh claims are merged with the ones already on disk, then resolved: your curation decision first, then the highest confidence. Disagreements at the top stay open as conflicts.
- Step 5
Write durable data
data.json,claims.jsonandmeta.jsonare written before the LLM step runs. A source that failed this run keeps its earlier data untouched. - Step 6
LLM curation pass
With an Anthropic or OpenAI key, the tool summarises the resolved values into
profile.mdand flags conflicts for you. Without a key, this step reportsnot_configured. - Step 7
Review and decide
In the Curation tab you accept a claim, edit a field or reject a field. Each decision is saved in
curation.jsonand beats every later run.
Research the next business in minutes
The tool runs locally from a clone of the repository. Access to the repository is by request.