agentleFS
agiea-brainagiea-brain/agency-skills/skills/hunter-harris/linkedin-export-contact-import/SKILL.md

linkedin-export-contact-import

Skill1 starsChanged 16 days agoMITView on GitHub

Use this skill when you've downloaded your LinkedIn data export and want your connections, sent invites and DM history in your CRM or contact sheet. Produces upserted contacts plus dated touches, safe to re-run every week. "Import my LinkedIn export", "load my connections into the CRM", "who have I messaged on LinkedIn".

The file

We split this skill into the rules it gives an agent

7 rules. About 0 are common, 6 are distinctive, and 1 is very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈6

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 1

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  • could be automatic: a setting could enforce it
  1. Distinctive:

    No scraping and no API: this works on the export LinkedIn gives every member.

    line 17
  2. Very specific:

    Read Connections.csv from the line that starts First Name,Last Name,.

    line 23
  3. Distinctive:

    MUST dry run before writing, and keep each export folder intact.

    line 40
  4. Distinctive:

    MUST key on the profile URL slug.

    line 41
  5. Distinctive:

    NEVER treat a draft, spam or group message as a touch.

    line 42
  6. Distinctive:

    NEVER use an old export to downgrade data a newer one wrote.

    line 43
  7. Distinctive:

    NEVER upload the export to a third-party service to parse it.

    line 44

≈ “Common” is an estimate: at least 1 in 5 of the 364,980 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Applies when LinkedIn holds relationship history your CRM doesn't. Produces contacts from Connections and sent invitations, and one dated touch per one-to-one message, deduplicated so weekly re-imports add only what's new.

No scraping and no API: this works on the export LinkedIn gives every member.

The play

  1. Request the export from LinkedIn's data privacy settings. The basic export has Connections, Invitations and messages. The complete one adds comments, shares and reactions. A weekly rhythm that works: basic midweek, complete at the weekend.
  2. Unzip each export into its own dated folder. Never overwrite the previous one.
  3. Read Connections.csv from the line that starts First Name,Last Name,. LinkedIn puts notes above the header.
  4. Key every person on the /in/ part of their profile URL. A name-only match is a candidate: list it for the human to confirm.
  5. Contacts come from Connections, plus OUTGOING rows from Invitations.csv. Count and skip members whose fields are all blank (hidden or deactivated).
  6. From messages.csv, keep only one-to-one threads. Skip drafts, spam and group threads. Each message becomes a touch: date, person, channel, direction.
  7. Dry run first. Show the proposed creates, merges and updates; apply only after the human approves.
  8. Change a contact's status only from recent messages: within 60 days of the newest message in the export. Import exports in date order and never one older than the last. A two-year-old thread is history, not a live conversation.

The file-by-file quirks are in references/export-file-notes.md. Open it before writing the parser.

What good looks like

  • First check: a second run of the same export adds zero touches. If it adds any, the dedupe key is wrong.
  • The mediocre version imports everything as current. Old threads flip contacts to "in conversation", group chats become replies, and drafts you never sent count as messages.
  • The output is right when the counts reconcile: Connections rows equal new plus matched plus skipped-blank, and five spot-checked people show the right last message date.

Rules

  • MUST dry run before writing, and keep each export folder intact.
  • MUST key on the profile URL slug; NEVER merge two people on name alone when more than one contact has that name.
  • NEVER treat a draft, spam or group message as a touch.
  • NEVER use an old export to downgrade data a newer one wrote.
  • NEVER upload the export to a third-party service to parse it; it contains other people's messages.

Download now (ZIP)

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
What did you use it for? e.g. Reviewing pull requests in a Rails app

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.