agentsclimarketplace

Linkedin connection note

Skill faiz-sayyad/gtm-outreach-skills/skills/linkedin-connection-note

Write a LinkedIn connection request note under 300 characters, using one of 4 proven patterns tailored to context. Use whenever the user needs to send a connection request with a note, asks how to phrase a LinkedIn invite, wants to reach a 2nd or 3rd-degree connection on LinkedIn, or asks for a "connect message" or "invite note." Also use before any LinkedIn DM to a stranger where the user isn't sure if they should connect first vs. InMail.From its SKILL.md

Install
npx -y skills add faiz-sayyad/gtm-outreach-skills --skill linkedin-connection-note

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

2 things to look at

  • no licenseNo license file was found in the repository. Code published without one is not open source by default, so using it at work is a question for whoever answers licensing questions where you are.
  • 0 stars0 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

SKILL.md

7.1 KB, ~1.5k tokens by cl100k_base, as published. Nobody here has run it

LinkedIn Connection Note

Draft a LinkedIn connection request note. Strict 300-character limit. Load voice.md before drafting.

When this skill triggers

  • User wants to connect with someone on LinkedIn and asks for the invite note
  • User asks how to phrase a "connection request" or "LinkedIn invite"
  • User is preparing to reach a 2nd or 3rd-degree contact and needs to bridge via connection first
  • User asks whether to send a note at all (default: yes, always send a note — blank invites get rejected 3x more)

Character budget

Hard limit: 300 characters including spaces. LinkedIn cuts off at 300. Over-limit notes get truncated mid-sentence, which looks worse than no note at all.

Character count check every note before returning. If over 300, tighten and re-run.

The 4 patterns

Rotate across these based on context. Never send the same pattern twice to the same buyer profile in a 30-day window.

Pattern 1 — Curiosity (default when there's a public signal)

Formula: reference the signal in one sentence + one question or reason to connect.

Best for: LinkedIn post engagement, podcast appearance, conference talk.

Example:

"Your post about outbound tool sprawl matched exactly what we hit last year. Curious what your consolidation path looks like — would love to stay connected."

Pattern 2 — Value (default when you have something specific to offer)

Formula: name the shared context + name the specific thing you'd share.

Best for: after a warm-market signal, or when you have a genuinely useful resource.

Example:

"Watched a lot of Series A martech teams rebuild their outbound in the last 6 months — wrote up a 90-day audit template. Happy to share if useful. Sending over for that + to stay connected."

Pattern 3 — Warm reference (when there's a mutual connection or shared context)

Formula: name the connection or context + why you're reaching out now.

Best for: mutual customer, mutual investor, mutual community, same alma mater or event.

Example:

"Sam [mutual connection] mentioned you were building out the sales team at Acme. Would love to connect — happy to trade notes on GTM engineering when timing's right."

Pattern 4 — Question (for research-mode outreach with no ask)

Formula: one narrow question tied to their public work, no product mention.

Best for: research outreach, thought-leader engagement, no immediate commercial intent.

Example:

"Read your Substack piece on signal decay — curious whether you'd draw the same curves for outbound in enterprise vs. SMB. Would love to connect and learn more."

Rules

Never do these:

  • Use the word "opportunity" (screams sales)
  • Say "I think we could help you" (they don't know you exist)
  • Reference your product or company name in the note (do that after connection is accepted)
  • Compliment their profile ("Impressive background", "Love your work")
  • Say "hope this finds you well"
  • Reference the platform ("Saw you on LinkedIn")
  • Use exclamation points
  • Ask for a call/meeting in the note itself

Always do:

  • Reference something specific and public about them (post, quote, work)
  • Keep it under 300 characters
  • Sound like you'd actually say the sentence out loud
  • End with a reason to connect that's low-friction ("Would love to stay connected", "Sending over to trade notes when timing's right")

Output format

Return in this structure:

CONNECTION NOTE — [Person] @ [Company]

Pattern: [Curiosity / Value / Warm Reference / Question]
Character count: [X of 300]

"[the note itself]"

Rationale: [1-line why this pattern fits this person]

If the user provides multiple people or wants variants, provide 2-3 versions across different patterns and let them pick.

Examples

Example 1 — LinkedIn post signal

Input:

  • Person: Sarah Chen, VP Sales at Acme
  • Signal: Her post about outbound tool sprawl
  • Context: She's a 2nd-degree connection, no mutual introducer

Output:

CONNECTION NOTE — Sarah Chen @ Acme

Pattern: Curiosity
Character count: 227 of 300

"Your post about 4 outbound tools not talking to each other hit — we ran that same Frankenstein stack for a year and a half. Curious how you're thinking about consolidation vs. patching. Would love to stay connected."

Rationale: Curiosity fits when there's a fresh public signal (her post) and you can credibly reference having lived the same problem.

Example 2 — champion job change

Input:

  • Person: Marcus Reid, new Head of Growth at Vector
  • Signal: Was at Beta (former customer), now at Vector
  • Context: Marcus already knows the user from Beta days

Output:

CONNECTION NOTE — Marcus Reid @ Vector.io

Pattern: Warm Reference
Character count: 178 of 300

"Congrats on the Vector move, Marcus. Was thinking about the outbound engine we shipped at Beta last summer — curious how much of that framework you'll bring here. Reconnecting."

Rationale: Warm reference is the obvious fit — you already have shared history. Note keeps it about the WORK, not about pitching the product to him at his new role.

Example 3 — cold research outreach

Input:

  • Person: Alex Vargas, GTM writer with a Substack about signal-based sales
  • Signal: His most recent piece on signal decay
  • Context: No prior relationship, user wants to engage as a peer, not a customer prospect

Output:

CONNECTION NOTE — Alex Vargas

Pattern: Question
Character count: 265 of 300

"Read your latest piece on signal decay — the curve you drew for enterprise felt right. Curious whether you'd draw it the same way for PLG-motion companies or if there's a different shape there. Would love to connect and trade notes."

Rationale: Question pattern fits peer/thought-leader outreach. Zero pitch. Grounds the connection in a specific idea from his work.

Common failure modes

Failure: note exceeds 300 characters. Fix: Count characters before returning. If over, drop adjectives, shorten sentences, remove throat-clearing phrases like "just wanted to say."

Failure: note reads like a sales pitch even though it's a "connection request." Fix: If the note mentions your product, company name, or an offering, remove it. Save that for the DM after they accept.

Failure: note is so generic it could be sent to anyone. Fix: The specific personalization detail must appear in the note. If the note works when you swap out the name and company for someone else's, it's too generic.

Failure: note uses one of the banned phrases and slips past. Fix: Do a final scan before returning. Banned phrases have a way of showing up when you're rushed.

Related skills

  • After connection is accepted, use linkedin-opener for the follow-up DM
  • If no response after 5–7 days, use follow-up-sequence

What ships with it

Read from the repository

Just SKILL.md. No reference files, no scripts.

Keep looking

Skills are one crate of 325,949. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.