Describe the person, not the filters
“Heads of operations at logistics companies in the Lyon area” is a sentence; in the interface it is four filters and a guess at the industry taxonomy. The agent turns the sentence into a search, runs it on your account and returns structured results — and because they are structured, it can then rank or group them, which the interface cannot do at all.
“Find heads of operations at logistics companies near Lyon.”
“Of these results, who has changed job in the last year?”
Your search, your account, your visibility
Results depend on the account running them: your network, your region, your seat. This is not a public index — it is what you would see logged in, which is both the value and the limit. Two colleagues running the same query get different lists, and that is normal.
Search results are not profiles
A result carries a display name and a public identifier, not the full profile. Reading experience, skills or contact details is a separate read on the person, which the agent does for the shortlist rather than for all two hundred results — reads count against the daily budget, so narrowing first is not pedantry, it is what makes the pass finish.
“Read the top five and tell me which ones actually run a warehouse.”
Three searches, not one
Classic, Sales Navigator and Recruiter each have their own search with their own filters and their own ranking. They are distinct tools here, so a shortlist always comes from a known search rather than from whichever one happened to answer. An agent that silently ran Classic while you assumed Sales Navigator gives you a shorter list and no warning.
Replay a search you already built
If you have spent twenty minutes building a search in the interface, paste its URL: it is replayed exactly, filters intact. Far more reliable than describing every criterion again and hoping the agent rebuilds it.
“Run this search and tell me who we already spoke to: <paste the URL>”
Common questions
Can ChatGPT search LinkedIn?
Not by browsing — LinkedIn blocks that. With your account connected, it runs the search on your own account and returns structured results it can then rank, filter or compare.
Why are the results different from what I see?
Search depends on the account: network, region, seat and premium context all change it. The agent sees what you would see logged in, no more.
Can it search companies and jobs too?
Yes, people, companies, posts and job postings each have their own search, and Sales Navigator and Recruiter add theirs when the account holds the seat.
Can it enrich the results?
It can read the profiles of a shortlist to add experience or contact details. Doing it for every result burns the daily action budget, so narrowing first is the right order.
How LinkedIn works with Nilyo
Connection. Secure hosted sign-in (your LinkedIn credentials go to LinkedIn, never to the agent). Sales Navigator and Recruiter are detected automatically when your account has them.
What your agent can do. People and company search, profile lookup from a name or URL, your network and invitations, private conversations (classic, Sales Navigator, Recruiter inboxes), posts, comments and reactions, job postings and applicants, Recruiter projects.
Pacing and safety. About 100 actions per day per account, profile views and searches included; calls are serialized so an agent can never run LinkedIn requests in parallel.
- Profile URLs are resolved to stable profile IDs before any action.
- Invitations carry an optional note of up to 300 characters.
- Private or restricted profiles return an explicit “not accessible” result instead of a retry loop.
All LinkedIn guides
Related guides
Give Nilyo to your agent.
Paste one setup link. Your agent can guide OAuth, account connection and the first useful action.