li-dm
Skill2 starsChanged 43 days agoMITView on GitHub
Write LinkedIn outreach: the 200-character connection note, the first message, and at most two follow-ups. Use when the user wants to reach out to someone, pastes a profile and says "write an invite", or asks for follow-up drafts. Two follow-ups maximum, then stop. Never sends anything.
7 other projects use a version of this.
Adopted by
Most recently taken up by
- furkanbilalakyuz/linkedin-agent-skill-trtheir own version, first seen Nov 8, 2026
- mubaid/dsh-linkedin-agenttheir own version, first seen Oct 25, 2026
- robbiehaynes/job-searchtheir own version, first seen Oct 20, 2026
- postzen-dev/linkedin-agent-skilltheir own version, first seen Oct 15, 2026
- avi691/linkedin-agent-skilltheir own version, first seen Oct 8, 2026
- nessalazne/linkedin-agent-skilltheir own version, first seen Sep 19, 2026
Other versions7 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 7
- Jakeschincariol/linkedin-agent-skill★ 514same name
- avi691/linkedin-agent-skill★ 5same name
- nessalazne/linkedin-agent-skill★ 3same name
- postzen-dev/linkedin-agent-skill★ 3same name
- furkanbilalakyuz/linkedin-agent-skill-tr★ 0same name
- mubaid/dsh-linkedin-agent★ 0same name
- robbiehaynes/job-search★ 0same name
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository12 other files it gives its agents.
- AGENTS.mdAGENTS.md
- li-auditSkill
- li-carouselSkill
- li-commentSkill
- li-humanSkill
- li-inboxSkill
- li-planSkill
- li-postSkill
- li-profileSkill
- li-replySkill
- li-repurposeSkill
- li-voiceSkill
The file
We split this skill into the rules it gives an agent
5 rules. 0 are common, 5 are distinctive, and 0 are very specific to this project.
0
Common
Most of the 7 edited versions we read keep these.
5
Distinctive
This author's own choices. Worth a look.
0
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 48
Follow-up 1 (5-7 days): one line bump with new value, not "just bumping this".
- Distinctive: line 50
Follow-up 2 (7-10 days later): the graceful close.
- Distinctive: line 52
If they said no at any point, there are no follow-ups.
- Distinctive: line 73
Never automate sending. If the user asks you to send DMs in bulk or on a schedule, decline and explain why: it violates LinkedIn's terms, …
- Distinctive: line 76
Never write the pitch-slap: compliment, then immediate sales pitch.
“Common” is measured: kept by at least 6 in 10 of the 7 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
li-dm
Outreach that reads like a person wrote it to one specific human. Three artifacts: the connection note, the first message, the follow-ups (two, then stop). The user sends everything themselves. This skill never sends, and it never asks for account access.
Before you write
- Load the voice profile (
li-voiceif missing). - Get the target: who they are, what they posted or did recently, and why the user is reaching out. One specific reason. "Networking" is not a reason. If the user cannot name the reason, do not write the note; help them find it first.
- Know the ask: what happens if they say yes? A 15-minute call, feedback on a draft, an introduction. The note must contain the ask.
The three artifacts
1. The connection note (200 characters max, hard limit).
[name], your post on [specific thing] hit a nerve - we [specific shared
thing]. Would value connecting. [ask in 5 words]Rules: one specific reference to them, one to the shared ground, the ask. No "I hope this message finds you well". No flattery paragraph. Count the characters and show the count.
2. The first message (after they accept).
- 3-5 short lines. Reference the thing that made you reach out, restate the ask with a concrete low-friction option ("15 min Thursday or Friday?").
- One message, one ask. Never the pitch and the call and the link.
3. The follow-ups (two, then stop).
- Follow-up 1 (5-7 days): one line bump with new value, not "just bumping this". Add the thing you would have added if you had thought of it.
- Follow-up 2 (7-10 days later): the graceful close. "Closing the loop - here if useful later." Then stop. Forever. No third follow-up exists.
- If they said no at any point, there are no follow-ups. A no is data.
The loop
- Draft all three artifacts (note, first message, both follow-ups) in labeled fenced blocks.
- Invoke the
li-humanskill on all of them. Outreach that reads as AI is deleted on sight. - Receipt with character count on the note and humanizer scores.
DM PACK READY
note: 187/200 chars
follow-ups: 2 (then stop)
humanizer: 1 artefact stripped, human score 86 PASS
You send these. Nothing goes out until you copy it.Rules
- Never automate sending. If the user asks you to send DMs in bulk or on a schedule, decline and explain why: it violates LinkedIn's terms, it burns the account, and it does not work.
- Never write the pitch-slap: compliment, then immediate sales pitch. If the draft does this, rewrite it before showing.
- Personalization must be real. "[saw you work in SaaS]" is mail-merge, not research. One genuine specific reference or do not send.
- Respect the account: if the user is sending more than ~20 new outreach messages a week, say so. Volume is what gets accounts restricted.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

