The candidates who matter do not apply, they announce
A developer finishing a contract, a marketing director leaving after a reorganisation, a salesperson "listening to the market": these people do not send CVs. They publish a post, often cautious, sometimes one line long, and their network does the rest. Three offers arrive that day. The next day, the conversation is already under way with someone else.
The "Open to work" banner does not tell the whole story: many never turn it on, out of discretion towards their current employer or simply out of habit. What they do is write. "Available from", "end of contract", "leaving", "open to opportunities", "next chapter" — the phrases are few and always come back.
Prosko watches those phrases, crossed with your roles, in the LinkedIn feed and search, and sends you every post that contains them: the person, their job title, the excerpt, the direct link. You see the profile while it is still free.
The rule that catches these posts
For a recruiter, the rule crosses two things: an availability phrase and a role. Either one alone surfaces too much.
The exclusions do the real work here. "Open to opportunities" also appears in half the job ads ("open to profiles…"); "we are hiring" and "we are looking for" discard them in one go. A multi-word trigger fires when all of its words appear in the post, in any order; case, accents and hyphens are ignored.
The role and the location are better handled by the intent given to the AI assistant: "someone announcing their own availability for a developer position, in the Lyon area or remote". The assistant re-reads the posts your words picked up and discards those that do not match — the job ad that slipped through, the outplacement firm's post about "supporting transitions", the person who is available but in a completely different field. That is what makes the rule liveable day to day: a few alerts a day, almost all relevant.
One rule per role, or per family of roles. Every alert carries the name of its rule; if several recruiters share the channel, each knows at a glance what concerns them.
What it catches
Two posts like the ones a rule of this kind surfaces in an ordinary week. The names are made up; the phrasing is not.
"End of contract on the 30th with my current client after eighteen months. Looking at what's out there for September: Node or Go, product team, Lyon or remote. DMs open."
"After four years building Vela's engineering team, I'm handing over at the end of September. A next chapter to write — taking the time to talk with people who build useful things."
The first is a dated availability, with the scope spelled out. The second is more muted — a CTO does not write "I am looking for a job" — but says exactly the same thing to anyone who can read. In both cases, the person is in no database, and may never be.
The alert you receive
"End of contract on the 30th with my current client after eighteen months. Looking at what's out there for September: Node or Go, product team, Lyon or remote. DMs open."
The person, their job title as shown on LinkedIn, the excerpt where the phrase appears, the link to the post. "Checked by the assistant" means the assistant re-read the post and confirmed it matches the intent you described. Alerts pile up in the extension — the last five hundred — and, if you wish, in a Slack or Discord channel shared by the recruiting team.
Reach out without scaring off
Reply the same day, not the same week. The person posted to attract offers; the first ones get real attention, the later ones a polite reply. Timing matters more than length.
Read the post before writing. Karim wrote "Node or Go, product team, Lyon or remote". A message that opens with "I saw you're looking for Node, product team, in Lyon" gets read; a generic message about "a great opportunity" does not. The alert's excerpt gives you everything you need to be precise.
Comment in public, offer in private. A short, sincere comment under the post — without pasting your job ad — shows you have read it, and the person's network sees it. The offer itself goes by private message.
Match the tone of the post. A muted announcement calls for a muted approach: "when you have a moment, I'd like to tell you about what we're building" beats a job description as an attachment.
What to know before you start
- Your LinkedIn account is enough. Recruiter, Recruiter Lite or an ordinary account: Prosko reads the feed and the search as your account displays them. It does not replace your sourcing tool; it flags what sourcing does not see — people who updated nothing, but wrote it down.
- It does nothing on your behalf. No messages sent, no connection requests, no automatic comments. Prosko tells you; the approach stays human, and your account stays yours.
- Everything runs in your browser. No server, no account on our side: your rules and finds never leave your machine. In return, monitoring works while Chrome is open.
- It reads posts. That is where availability gets announced; the rest of the profile is one click away from the alert.
Frequently asked questions
Does it detect the "Open to work" banner?
No, and that is not the point: the banner is already filterable in sourcing tools. Prosko catches what those tools do not see — people who turn nothing on but write that they are available, or about to be.
How do I target a role or a location?
Through words and intent. Add the role's vocabulary to the triggers if you like, but above all describe the profile in one sentence in the intent given to the AI assistant: "a developer announcing their availability, in Lyon or remote". It discards what does not match.
How many alerts a day should I expect?
With precise triggers, exclusions and a well-described intent, a few a day for a given role. If you receive many more, tighten up: it means exclusions are missing or the intent is too vague.
Can several recruiters share the alerts?
Yes, by sending alerts to a shared Slack or Discord channel. Each person running the extension needs their own licence; the channel is read by the whole team, and every alert carries the name of the rule that produced it.
Do candidates know they were spotted by a tool?
No. Prosko interacts with nobody: it reads what LinkedIn displays and passes it on to you. What the person sees is your comment or your message — written by you.
Page updated on .