agentleFS
sloemo01sloemo01/hermes-skills-bundle/linkedin-automation/SKILL.md

linkedin-automation

Skill9 starsChanged 4 months agoMITView on GitHub

LinkedIn people search via Kimi WebBridge — URLs, filters.

  • Sends data out

2 other projects use a version of this.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

3 rules. About 0 are common, 1 is distinctive, and 2 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈1

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 2

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Use with origin=FACETED_SEARCH parameter.

    line 42
  2. Very specific:

    Search school name alone: keywords=Amrita%20Vidyalayam

    line 52
  3. Distinctive:

    If no results, try variations: "Amritha Vidyalayam", "Amrita Vidyalaya", "Amrita Vidhyalayam"

    line 55

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 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

LinkedIn Automation with Kimi WebBridge

Patterns for finding people on LinkedIn via browser automation. Use when you need to search for profiles, filter by connection degree, paginate results, or search by education/employment.

Quick Reference: Search URL Patterns

GoalURL Pattern
Basic name searchhttps://www.linkedin.com/search/results/people/?keywords=NAME
1st-degree connections only...&origin=FACETED_SEARCH&network=%5B%22F%22%5D
2nd-degree connections only...&origin=FACETED_SEARCH&network=%5B%22S%22%5D
3rd+ connections only...&origin=FACETED_SEARCH&network=%5B%22O%22%5D
Specific person's connectionshttps://www.linkedin.com/in/USERNAME/connections/
Search within connectionshttps://www.linkedin.com/search/results/people/?keywords=TERM&origin=MEMBER_PROFILE_CANNED_SEARCH
Education/school searchhttps://www.linkedin.com/search/results/people/?keywords=SCHOOL_NAME
Name + school combohttps://www.linkedin.com/search/results/people/?keywords=NAME+SCHOOL
PaginationAdd &page=N to any search URL

Workflow

  1. Navigate to search URL with newTab:true and a descriptive group_title
  2. Snapshot the page (full:true for complete tree, full:false for interactive elements)
  3. Extract profile links from @e refs in the snapshot (look for role: "link" with profile names)
  4. Click a profile link to open it, then snapshot for details
  5. Paginate by clicking page number buttons (@e refs for "Page 1", "Page 2", etc.) or navigating to &page=N

Connection Degree Filters

Network ValueMeaning
%5B%22F%22%5D1st connections (encoded ["F"])
%5B%22S%22%5D2nd connections (encoded ["S"])
%5B%22O%22%5D3rd+ connections (encoded ["O"])

Use with origin=FACETED_SEARCH parameter.

Profile URL Pattern

Individual profiles follow: https://www.linkedin.com/in/USERNAME/

From search results, click the profile name link (@e ref) to navigate, then read the URL from the snapshot's url field.

Education Search Tips

  • Search school name alone: keywords=Amrita%20Vidyalayam
  • Combine with name: keywords=Arjun%20Kannan%20Amrita%20Vidyalayam
  • Filter by 1st/2nd connections to find alumni in your network
  • If no results, try variations: "Amritha Vidyalayam", "Amrita Vidyalaya", "Amrita Vidhyalayam"

Pitfalls

  • LinkedIn blocks automation — some actions may fail with "event.isTrusted" errors. Manual interaction needed for captchas, login walls.
  • Search results are truncated in snapshots — use full:true and parse the tree, or click through pages.
  • Profile URLs change — usernames may differ from display names. Always capture the actual URL after clicking.
  • Rate limiting — LinkedIn may throttle rapid requests. Add delays between actions.
  • Dynamic content — @e refs change on each snapshot. Re-snapshot after navigation before clicking.

Example Session Flow

# 1. Open search tab
curl -X POST http://127.0.0.1:10086/command \
  -d '{"action":"navigate","args":{"url":"https://www.linkedin.com/search/results/people/?keywords=Arjun%20Kannan","newTab":true,"group_title":"Arjun Kannan search"},"session":"linkedin"}'

# 2. Get full page tree
curl -X POST http://127.0.0.1:10086/command \
  -d '{"action":"snapshot","args":{"full":true},"session":"linkedin"}'

# 3. Click a profile (use @e ref from snapshot)
curl -X POST http://127.0.0.1:10086/command \
  -d '{"action":"click","args":{"selector":"@e33"},"session":"linkedin"}'

# 4. Get profile details
curl -X POST http://127.0.0.1:10086/command \
  -d '{"action":"snapshot","args":{"full":true},"session":"linkedin"}'

# 5. Filter to 1st connections only
curl -X POST http://127.0.0.1:10086/command \
  -d '{"action":"navigate","args":{"url":"https://www.linkedin.com/search/results/people/?keywords=Arjun%20Kannan&origin=FACETED_SEARCH&network=%5B%22F%22%5D","newTab":true},"session":"linkedin"}'

Related Skills

  • kimi-webbridge — underlying browser automation tool
  • computer-use — alternative desktop automation approach

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.