French company dataConnector
INSEE Sirene
Search French companies by activity, department, headcount or revenue and resolve any SIREN or SIRET, from Claude or ChatGPT, through Tulina's own MCP server.
Official site: www.sirene.fr
What it can do
4 read/1 write
- 01Read
Search companies by activity code, department, postcode, headcount band, size category, revenue range or collective agreement
- 02Read
Resolve a SIREN to full identity, or a SIRET to a specific establishment
- 03Read
Find a company's registered headquarters among its establishments
- 04Read
Produce the avis de situation for an establishment
- 05Write
Enumerate exhaustively from the SIRENE stock when a result set runs past the search API's ceiling
Two things about Sirene cost people real work.
Geographic filters match any establishment, not the head office. Ask for companies in département 75 and you also get the ones headquartered in Lyon with a Paris office. When you mean the siège, you have to say so.
Enumeration truncates silently past roughly ten thousand results. No warning, no flag, just a shorter list than reality — a market map that looks complete and is not. Past that point the connector reads the Sirene stock, which has no cap.
How do you connect INSEE Sirene to Claude and ChatGPT?
Through Tulina's own MCP server. Search by activity code, department, postcode, headcount band, size category, revenue range or collective agreement, resolve a SIREN to its full identity or a SIRET to one establishment, find a company's registered headquarters, produce an avis de situation, or enumerate past the search API's ceiling from the Sirene stock itself — from the same chat as Claude or ChatGPT.
Why through Tulina
The connection was never the hard part.
Most systems now publish their own way for a model to reach them, and that is good: it is what Tulina connects to. The hard part is deciding who may use it, what it is allowed to return, and what the model knows before it asks.
The keys stay in one place
YouTeamOrganisation
Connect INSEE Sirene once. The credential is encrypted in a vault, never held in clear, and never pasted into a chat or a workflow file. A person uses their own key where they have one, their team's where they do not, the organisation's after that, so nobody is blocked and nobody holds more access than they need. Revoke it once and everything running on it stops.
What leaves is filtered before it leaves
fields droppedmasked or hashedpseudonymised
Roles decide who can reach INSEE Sirene at all. Beyond that, a policy on the connector itself decides what comes back. The data is minimised at the source rather than after it has already reached a model.
Every call is on the record
whatwhowhen
Runs are logged for the organisation. That is what makes an automated action reviewable after the fact, and it is what a security review asks for first.
The run starts knowing your company
namingkey accountsdecisions
A direct connection gives a model access. It does not tell it how your company names things, which accounts matter, or what was decided last quarter. Tulina carries that across every tool, so the same request against INSEE Sirene returns something usable instead of something generic.
Processes that run on INSEE Sirene
Find doctors and clinics
A census of every doctor and clinic in a French city, from map listings cross-checked against the practitioner registry and facility directory, with gaps named.
6 steps →Outreach based on government data
Every candidate's website is verified before a credit is spent, tiered by signal strength, then written one grounded opener that is checked for repetition.
7 steps →Qualify inbound leads
AI lead scoring for inbound leads: each one matched to its legal entity, scored on real headcount and filed accounts, and written back with the reason.
5 steps →Reply to every LinkedIn conversation
Four times a day, LinkedIn threads where the other person wrote last come back to you in Slack with a reply drafted in your style. Nothing is sent for you.
9 steps →
