Claude for LinkedIn Networking for Recent Graduates: 28 Claude Skills

What these skills are

A graduate's first month on LinkedIn, as installable Claude skills.

This is Claude for the first month on LinkedIn and the weekly rhythm after it: your story found, the profile built from it, one project turned into proof of work, posts and comments that sound like you, recruiters and alumni able to find you — and twenty minutes a week that keeps it alive.

The skills follow that month in the order it happens, in seven stages, and each stage hands a named artifact to the next. Know your story hands over the experience inventory, the target role brief, the voice guide and your AI use policy. Build the profile turns those into the headline, the About section, the experience, education and projects entries, the skills list, and an audit of the whole thing. Show your proof of work hands over the project interview notes, the one-page case, the Featured plan and the recommendation request. Post as a graduate hands over the ideas bank and the learning, project and milestone posts, with the voice check over all of them. Comment and connect hands over the follow list, the comment angles and the connection note. Be found hands over the Open to Work decision, the recruiter reply plan and the alumni search plan. The weekly 20 minutes hands over the routine, the activity log and the monthly refresh — which adds the month's new evidence back to the inventory and starts the loop again.

The red line: you did the work, Claude helps you say it. It never touches your account — you write, post, comment and connect yourself.

Each skill is one named method or one artifact: the inputs it needs, where it breaks, a fixed output template, a finish line and the next skill to run. No invented facts — never a project, a number, a skill or a quote you cannot back up; a missing figure becomes a bracketed placeholder, and a gap is marked as a gap. Nothing grades, ranks or profiles a person: the skills check your profile, your drafts and the fit to your brief, never you or the people you contact.

Where a skill fits a Claude surface — a Project to hold your inventory and notes, memory with editable Topics for your voice guide, Claude Docs for the project case, a Google Sheet for the activity log — it says so, and beta features are marked beta. Claude never opens, browses or acts on your LinkedIn account: you paste text in, and you do everything on LinkedIn yourself.

Mechanically, each skill is one folder with a SKILL.md file. The 28 are grouped into the seven stages, and each one gives full value on its own: run one, run a stage, or work the month through in the order it arrives.

Download all 28 skills. One zip: ready-to-install skill zips, readable SKILL.md files, and the sources behind every method. Free, no signup. The same 28 skills are open on GitHub: polar-bear-org/claude-skills.

The 28 skills you get

The set follows the first month on LinkedIn, in seven stages: knowing your story, building the profile, showing your proof of work, posting as a graduate, commenting and connecting, being found by recruiters and alumni — and the weekly 20 minutes that keeps it going.

1 · Know your story

1. Experience Inventory

Use when: You think you have no real experience and the profile looks empty

Output: Interview notes, one evidence row per real thing you did, strengths backed by two rows, gaps marked as gaps

2. Target Role Brief

Use when: You are about to write a headline and cannot say what you want next

Output: Two or three target roles, employer words from five real job ads, matching evidence rows, what to say yes to next

3. Personal Voice Guide

Use when: Every AI draft sounds like a stranger, and like everyone else's

Output: Your words and phrases, sentence length, words you never use, AI tells to strip, before and after lines

4. Personal AI Use Policy

Use when: You want a straight answer ready when an employer asks how you use AI

Output: One page on what you use Claude for and never do, how you disclose AI help, your check before anything goes public

2 · Build the profile

5. LinkedIn Headline

Use when: Your headline says “Student at [university]” and nothing else

Output: Three headline options within LinkedIn's limit, status plus target role plus proof, the evidence row behind each

6. LinkedIn About Section

Use when: The About box is empty, or full of “passionate, driven, motivated”

Output: One or two short paragraphs in your voice, every claim tied to an evidence row

7. Experience Entries

Use when: You think a bar job or a society role does not count

Output: Two to four CAR lines per role, numbers only with a source, transferable skill per line

8. Education and Projects Entries

Use when: Your strongest evidence is coursework and it sits as one line

Output: Degree entry with matching modules and dissertation, Projects entries with your part, links as media

9. LinkedIn Skills Section

Use when: You added 40 skills and none is the one the job ad asks for

Output: Five to ten skills from your brief, each linked to proof, a “learning” list for the rest

10. LinkedIn Profile Audit

Use when: The profile is “done” and you want to know what a reader sees first

Output: Section-by-section check, truth check against your CV and evidence, fixes in order

3 · Show your proof of work

11. Project Interview

Use when: You did a real project and cannot get past “we got a good mark”

Output: Six-moment interview on one project, your words kept, a do-not-claim list

12. Project Case Study

Use when: You want one thing a recruiter can open that shows how you think

Output: One-page case, a 150-word Featured version, the permission note

13. Featured Section Plan

Use when: Featured is empty, or holds work you are not sure you may share

Output: Three to five items to pin, a line for each, permission check, reading order

14. LinkedIn Recommendation Request

Use when: You want a recommendation without the default request or writing it for them

Output: Who to ask, a short personal request with real reminders, a thank-you

4 · Post as a graduate

15. Post Ideas Bank

Use when: You want to post but have “nothing to say yet”

Output: Ten ideas from your real month, the one true thing each says, five you will never post

16. Learning Post

Use when: You went to an employer talk or finished a course and want to say something real

Output: One post in What? So what? Now what? order, first line inside the moment

17. Project Post

Use when: You have a project case and people will not open Featured

Output: One post on one project moment, first person, optional carousel outline in text

18. Milestone Post

Use when: You got the offer and the usual announcement post makes you wince

Output: Short post for an offer, placement or degree, bottom line first, thanks only with a yes

19. Voice Check

Use when: You used AI on a draft and need to know it still sounds like you

Output: Draft checked against your voice guide, AI tells flagged, claims matched to evidence, every change shown

5 · Comment and connect

20. Follow List

Use when: Your feed is noise and nothing on it relates to the job you want

Output: Up to 20 employers, alumni and bodies you chose by hand, why each, what to watch for

21. Thoughtful Comment

Use when: “Great post!” is all that comes to mind

Output: Two or three angles you could genuinely add to one post, the facts you need, no flattery

22. Connection Request Note

Use when: The default request to someone you met feels like spam

Output: Short personal note within LinkedIn's limit, no job ask, a follow-up line for later

6 · Be found by recruiters and alumni

23. Open to Work Settings

Use when: You are not sure the green frame helps or looks desperate

Output: Which visibility fits your situation, job preferences from your brief, the privacy caveat

24. Recruiter Reply

Use when: A recruiter messaged you and you do not know if it is real

Output: Scam check first, questions to ask, a reply in your words, what to log

25. Alumni Search Plan

Use when: Alumni in your target roles exist and you do not know where to start

Output: Manual filters on your university page, up to ten alumni you choose, the reason and one question each

7 · The weekly 20 minutes

26. Weekly LinkedIn Routine

Use when: You post three times after a rejection, then disappear for a month

Output: A 20-minute weekly plan, an if-then plan for a rejection week, a stop rule for scrolling

27. LinkedIn Activity Log

Use when: You cannot remember who you spoke to or which posts started conversations

Output: One row per action, what it led to, notes for the monthly refresh

28. Monthly Profile Refresh

Use when: Your profile still describes you on the day you made it

Output: New evidence added, profile lines to update, one thing to stop and one to try

Where to start, and how the skills chain

There is no compulsory sequence, but there is a first skill: glink-experience-inventory. Almost everything after it reads its evidence rows, so if the profile skills feel like guesswork, that is the one that is missing. Every skill lists the inputs it needs, so it tells you when the stage before it is the one to run.

They chain. glink-experience-inventory finds the evidence, glink-target-role-brief says what it is for, glink-personal-voice-guide says how it sounds and glink-ai-use-policy gives you a straight answer on how you use Claude. From those four, glink-linkedin-headline, glink-about-section, glink-experience-entries, glink-education-projects and glink-skills-section write the profile line by line, and glink-profile-audit reads the whole thing back. glink-project-interview gets the real story of one project, glink-project-case turns it into a page a recruiter can open, glink-featured-section decides what to pin and glink-recommendation-request asks the person who watched you do it. glink-post-ideas-bank fills the month, glink-learning-post, glink-project-post and glink-milestone-post draft the three kinds you will actually write, and glink-voice-check holds every draft against your voice guide. glink-follow-list makes the feed worth opening, glink-thoughtful-comment finds something real to say on it, and glink-connection-note turns a conversation into a connection. glink-open-to-work, glink-recruiter-reply and glink-alumni-search-plan handle being found. Then glink-weekly-routine, glink-activity-log and glink-monthly-refresh keep it running — and the refresh feeds the month's new evidence back into the inventory.

The sources behind each method are in resources/evidence-and-sources.md, inside the pack.

Setup guide

1. Download the pack. One zip: an install folder with 28 ready-to-upload skill zips, a skills folder with the same 28 skills as readable SKILL.md files, and resources/evidence-and-sources.md, which names the source behind every method.
2. Install your skills. In Claude Code, add the marketplace and install grad-linkedin-pack, and all 28 load at once. In Claude, turn on code execution in Settings, Capabilities, then go to Customize, Skills and upload one zip per skill from the install folder. Prefer working from files? Add the SKILL.md files to your Project knowledge instead; it works, just less cleanly.
3. Start with your story, not your headline. Run Experience Inventory first: it interviews you one question at a time and writes the evidence rows every other skill reads. Add the Target Role Brief and the Personal Voice Guide, and the profile skills have what they need. Each skill lists the inputs it needs and names the next skill to run, so one run turns into the chain on its own.

Where to start

The quality bar

Every skill in the pack holds the same standard, the one we hold when we write about our own work:

• Every line comes from your own interview notes or evidence rows: anything without a row is a gap, not a claim
• Numbers, skills and quotes appear only with a source you could show in an interview; otherwise they become bracketed placeholders, or they come out
• Drafts follow your Personal Voice Guide, with AI tells stripped and every change shown, and you decide what stays
• Nothing grades, ranks or profiles a person: the skills check your profile, your drafts and the fit to your brief, never you or the people you contact
• Confidential, assessed or other people's work is checked for permission first, and legal points end with “check with a qualified adviser”
• You did the work; Claude helps you say it: it never adds a project, a number, a skill or a quote you cannot back up
• It never touches your account: you write, post, comment and connect yourself, with no automation and no scraping

Who made this

Polar Bear is a people ops consultancy for human-size teams (20 to 200 people). Built by ex-McKinsey founders with a dream to make AI work for People, not instead of them. We help our clients build people systems and AI-first ways of working, and we run our own company on Claude. This pack is the free, self-serve version of how we work.

The pack carries a graduate's first month on LinkedIn. When you want your whole team working this way, AI carrying the overhead so people do the part only people can do, across hiring, early careers and everyday operations, that's what we build with clients.

Find 30 min together

Install in Claude Code

Two lines, and every skill in the pack loads at once.

/plugin marketplace add polar-bear-org/claude-skills
/plugin install grad-linkedin-pack@polar-bear-skills

Using Claude on the web instead? Download the zip and upload each skill from its install folder under Customize → Skills.