LinkedIn Warm Outbound
This week's warm leads: the people engaging with posts about your problem, scored and staged with a first message each.
It stages every campaign in your sequencer, stopped, and never starts one. It sends email only when you say so for that run, and replies only once you approve them. Its LinkedIn reads never use your account.
»Our own LinkedIn account, Jul 1 to Aug 24, 2026, this recipe. The dashboard is in the receipts below.
searches written from what you sell · buyer language, not vendor jargon
Our SDRs sent 200 cold emails a day. We cut it to 40 warm ones and doubled meetings booked.
214 reactions · 38 comments
Your next 20 customers already commented on someone else's post this week. You just aren't reading it.
1,904 reactions · 240 comments
Firmographics tell you who could buy. Engagement tells you who is thinking about it this week.
1,211 reactions · 156 comments
We stopped buying lists in March. Pipeline is up 40%. The list was the problem the whole time.
321 reactions · 55 comments
Reply rates on our engagement-sourced list: 34%. On the bought list: 3%. Same message. Same week.
402 reactions · 61 comments
We enriched 4,000 leads last quarter. We contacted 400. That's $1,800 of enrichment we lit on fire.
705 reactions · 97 comments
deduped at the source · ICP check runs before enrichment spends anything
HeyReach · Instantly · AgentMail · Gmail · into the one lane you picked
Saw you've been in a lot of the churn conversations lately. Most of it traces back to who the list was built from…
You've been deep in the outbound tooling threads. Honest answer on tools: it depends on where your list comes from…
Saw you've been in the AI SDR conversations. The sourcing side is the whole game, happy to show what we mean…
List quality keeps coming up in the threads you're in. Same scar here. We source from engagement now, want the breakdown?
Your market tells you who's shopping, in public, every day: they react to and comment on posts about the problem. This agent finds those posts, pulls everyone on them, and scores each person against your ICP before it pays for a single email lookup. The ones who fit land in HeyReach, Instantly, AgentMail or your Gmail with a first message written from what they engaged with. You press start.
One paste into your agent.
Your first warm lead list, minutes later.
Install superagnt's LinkedIn Warm Outbound skill for me. It's the warm-outbound-engine plugin from the github.com/superagnt/plugins marketplace, plus the superagnt MCP server (on Claude it's a connector I add in my browser; elsewhere I sign in through my browser). The setup guide for each kind of agent is at https://www.superagnt.com/install?skill=warm-outbound-engine» installs the warm-outbound-engine skill (read every file first, or the raw SKILL.md) and enables data:linkedin, database
We run this on ourselves.
Our own LinkedIn account, eight weeks, this exact recipe. We never measured a before for this account, so there's no before-and-after here. The numbers are read straight off the dashboard.
The post they engaged with predicts more than their score
We messaged 1,221 people this skill's way between May 3 and Oct 4, 2026, and matched each to what happened. The share tagged Interested doubled from the weakest source post to the strongest, while none of the four fit scores we tested told repliers from the rest (AUC 0.49 to 0.55, where 0.5 is a coin flip).
| Source post | Messaged | Replied | Tagged Interested |
|---|---|---|---|
| Founder-led sales, doing my own sales | 179 | 35% | 24% (18.4 to 30.8) |
| AI SDR, GTM AI agents | 68 | 43% | 24% |
| Cold email | 436 | 35% | 20% (16.5 to 24.0) |
| Cold calling | 232 | 31% | 16% |
| SDR and prospecting tips | 142 | 24% | 12% (7.6 to 18.3) |
So the skill fishes where buyers post about the problem, uses the score to drop the wrong people before it pays for an email, and lets replies steer which searches run next. Country barely moved the rate: 18% in the US, 20% in the UK, 18% in Canada.
»Our per-source yields study, Oct 4, 2026: 1,221 messaged LinkedIn engagers matched to outreach outcomes; ranges are 95% intervals. One sender, one offer.
From 28,077 engagers to 1,221 conversations
| Step | Count | Note |
|---|---|---|
| Posts scraped | 1,274 | Posts on the problem, May 3 to Oct 4, 2026 |
| Unique engagers | 28,077 | About 22 a post |
| Profiles read | 23,987 | 85% of engagers |
| Dropped | 10,263 | 9,514 of them for region |
| Sent to the sequencer | 3,677 | 13.1% of engagers |
| Messaged, matched to outcomes | 1,221 | 33.3% replied, 18.4% tagged Interested |
The data bill: profile reads alone came to about $216 (23,987 at 0.9 cents each), and with reactions, comments and searches it was at least $224, or 6.1 cents of data per lead sent and about 98 cents per Interested reply, before scoring. The skill reads each profile the same way, then scores each person before it pays for an email, and a LinkedIn-only lane skips the email spend entirely.
»The same study: our own outbound workspace, joined to the sequencer's per-lead outcomes.
Is it safe, and what are the limits?
- The playbook's ceiling, from its Sep 2026 research: about 200 connection requests a week per account, and an account that hasn't been sending gets locked if it suddenly sends 100 or more. Ramp a new sender slowly and let your sequencer's per-account limits set the daily pace.
- Requests go out blank by default: ours accepted at 62.3%. Custom notes are capped each month, so the skill saves them for the top 10 to 20% by score.
- Pending requests are withdrawn after about 10 days.
- Never rented or "avatar" profiles, resold Sales Navigator seats, or tools that get around LinkedIn's caps.
- LinkedIn's User Agreement (section 8.2, read Nov 3, 2026) bans "bots or other unauthorized automated methods" that add contacts or send messages. The skill's reads never sign in as you, and nothing sends until you start the campaign in your sequencer, which acts through the LinkedIn account you set up there.
»The skill's outreach playbook, distilled in September 2026 from 25 full videos by Alex Hormozi, Nick Abraham and Alex Berman, 60 of Nick Abraham's posts on X, and our own production outbound. It ships in the plugin below.
Two first messages, and the sequence around them
Examples: the playbook's templates, with the blanks the skill fills from each person's row.
- Example, for an engager: "Saw you're engaging with a ton of {topic} content on LinkedIn. Would love to {give value}. Open to it?"
- Example, with proof: "Hey {first_name}, saw {specific signal}. We help {type of company} {specific outcome}. Recently we helped {similar company} {specific result} in {timeframe}. Worth sending over {low-risk asset}?"
The sequence shape the skill checks your campaign against:
- LinkedIn: the opener; one line of proof 15 minutes later; a result plus a document 5 days later; a last short ask 4 days after that; then check-ins at 2 and 3 months as a separate campaign you start.
- Email: two or three touches over two to three weeks, each adding something new. Never "just bumping this".
- Any reply, a "not interested" or a late competitor match stops the sequence.
Seven steps to your first warm lead list.
- 1
Tell it who buys, once
Four answers: who owns the problem and who looks close but never buys, the competitors and creators your buyers follow, who it should never contact, and the one lane you send from. It drafts all four from your website, so you only fix what's wrong.
agnt_tools_enable · agnt_db_insert
icp_profile · drafted from your siteexamplebuyercategory + retail ops leads at CPG brands
never4 competitors · 3 customers · your team
watch2 competitors · 6 creators
laneheyreach
you correct it once
- 2
Fish where buyers post
Three ponds: posts from buyers who have the problem right now (hiring for it, breaking under it, shopping for a fix), the latest posts from your watchlist, and the topic itself. Searches use the words buyers post in, because vendor jargon only finds other vendors. No term repeats inside 14 days.
data_linkedin_search_posts · data_linkedin_search_post_by_hashtag · data_linkedin_get_company_s_post · data_linkedin_get_profile_s_posts
this run's searches, by pondexamplebuyers“hiring a category manager”40%watchlistRivalCo · 2 recent posts30%topic“shelf reset”30%» no term reused inside 14 days
- 3
Pull everyone on the post
The reactors, the commenters, and the author when a buyer wrote it. Each row keeps what that person did: the comment itself, or the reaction. Anyone already in your table is dropped at the source, and so are your competitors and their staff.
data_linkedin_get_post_reactions · data_linkedin_get_profile_post_comment · agnt_db_upsert
everyone on one postexample41 reactors · 12 commenters · 1 author» new · Dir. Category, Norrin · wrote the post
» new · Head of Retail, Mesa Co · commented
» new · VP Growth, Halloway · reacted
» dropped · 27 already in your table · 3 competitors
- 4
Score before you pay for an email
A narrow gate drops fake profiles, the wrong region and your do-not-contact list. Everyone left gets one score from 0 to 100 against your ICP, with the reason on the row. Only leads over the bar get an email found and verified, and a LinkedIn-only lane skips that spend entirely.
data_linkedin_get_profile_data_by_url · data_agnt_companies_enrich · data_agnt_people_email_finder · data_agnt_people_email_verifier
26 new leads · the score gates the spendexampleover 50 · email found + verified14under the bar · nothing spent9gated · wrong region372 · owns category reviews at a 40-store brand, hiring for it
- 5
Open on what they engaged with
One message with one job: a reply. It opens on the topic they keep engaging with, never the creator whose post they were on, and asks for something small. It can also offer a free resource built from their own company's data. You approve the first five before anything is staged.
agnt_db_upsert
first touch · opener: topicexample“Saw you've been in a lot of the category review threads lately. We put together a shelf-gap breakdown for your top SKUs. Want it?”
» the topic, never the creator · one small ask
you approve the first five
- 6
Stage it, never start it
A HeyReach list, an Instantly campaign, AgentMail drafts or Gmail drafts, whichever lane you picked. Each lead carries its own message, and the campaign stays stopped. Pressing start is yours, and it's the last point where a bad list can still be caught.
connection_heyreach_list_add_leads_v2 · connection_instantly_bulkaddleads · connection_agentmail_create_draft
staged in the lane you pickedexampleheyreachinstantlyagentmailgmail» 14 leads on the campaign's list, each with its own message
campaign stopped · you press start
- 7
Let replies steer the next run
Webhooks bring replies back onto the same rows, sorted by intent, with an answer drafted for you to approve. The next run writes more searches like the ones that got answers. Once a run looks right, the same steps go on a schedule as a small hosted agent team.
agnt_webhooks_create_endpoint · connection_heyreach_webhook_create · connection_instantly_createwebhook · agnt_db_execute_sql
replies, feeding the next runexamplewebhook · Mesa Co replied · interested» answer drafted · waiting on your ok
» “category review” · 31 leads · 9 replies
» “retail media network” · 22 leads · 0 replies
next run writes more like the first
»After the first good run, the skill offers the always-on version: a small hosted runner on a schedule, so the work keeps happening with your laptop closed. You confirm the cadence before anything is scheduled.
Read every file before you install it.
SKILL.md tells your agent to draft your buyer profile from your website, search the posts your buyers engage with, pull everyone on them, score each person before paying for an email, write one first message from what they engaged with, and stage the list stopped in the sequencer you pick. It opens with the outreach library it ships, then setup for the agent: which tools to turn on, and the tool names on newer servers.
Plain markdown your agent reads while it runs. SKILL.md is the procedure, and references/ holds the 5 files it opens when a step needs more depth. This is the plugin exactly as it installs from github.com/superagnt/plugins: 9 files, 113 KB.
Common questions
01Is it against LinkedIn's rules to send automated messages?
LinkedIn's User Agreement (section 8.2, as of Nov 3, 2026) bans bots and unauthorized automated methods that add contacts or send messages. This skill never starts a campaign: its LinkedIn reads never use your account, and the list lands stopped in your sequencer. A sequencer acts through the LinkedIn account you connect to it, so starting a campaign, and its risk, is your call.
02How many messages or connection requests a day or week are safe?
The playbook the skill follows: about 200 connection requests a week per account at most, and never 100 or more at once from an account that has been quiet. Ramp new senders slowly and let your sequencer's per-account limits set the daily pace.
03Can ChatGPT write LinkedIn messages?
Yes. Any capable model can write the words; what decides the reply is who gets the message and what it opens on. This skill runs in Claude, ChatGPT or Codex and writes each first message from what that person engaged with, on live LinkedIn data.
04Should AI reply to prospects automatically?
Not on its own. The skill brings replies back onto the same rows, sorted by intent, with an answer drafted for you to approve. A reply agent you deploy sends only what you approved, unless you choose to let one kind of reply go out by itself.
05What reply rate should I expect?
Ours, over 8 weeks on our own account: 62.3% of 1,582 requests accepted, and 36% of the 866 people messaged replied. On 1,221 people messaged from May to Oct 2026, 33.3% replied and 18.4% were tagged Interested, from 12% to 24% depending on the source post. Your offer and audience will move it.
06Does warm outreach beat cold DMs, and by how much?
We haven't run a cold-list control on the same account, so we don't claim a multiple. What we measured inside warm lists: the strongest source posts drew twice the Interested rate of the weakest, 24% against 12%, on 179 and 142 people.
Already connected? Your agent can list every packaged skill: agnt_tools_search + agnt_guidance_search