— Quick Read

Turn a LinkedIn Profile Into a Short Business Brief

This workflow creates an online agent that takes a LinkedIn profile URL and returns a short, evidence-backed person report. It is useful before a sales call, partnership conversation, recruiting screen, executive briefing, or networking meeting because it turns public profile and activity signals into a compact summary.

Want the full recipe? Here is the reusable guide an AI builder can follow.

You can also try the finished agent for yourself here.

What You Are Building

You are building a business research helper. Give it a LinkedIn person profile URL, and it collects public profile details and public posts, then creates a short report with a snapshot, activity signals, recurring themes, notable posts, and conversation starters.

The agent runs on Blocks, a network for running AI agents and connecting specialized agents to each other.

What the Tools Do

  • Blocks is the agent network. It lets this person-report agent be registered, called, and reused by other workflows instead of living only in one chat.
  • Railway is where the agent runs continuously online, like a small always-on worker.
  • Apify is the service that collects structured public information from LinkedIn profile and post pages.
  • GitHub is where the reusable instructions are stored so another AI builder can reproduce the workflow.
  • An API key or token is a private password-like credential that lets the agent use a service.
  • A CLI is a command-line tool the builder uses to register, deploy, and test the agent.
  • A dataset is the structured result returned by a research run, such as profile fields or recent posts.

This version does not require OpenAI because the short report is generated with deterministic TypeScript logic. A future version could add OpenAI if you want richer narrative analysis.

What the Guide Gives You

The linked guide gives an AI builder the practical recipe to:

  • Recreate the Node/TypeScript Blocks provider agent.
  • Configure Apify profile and activity actors.
  • Keep comments and reactions off by default to control cost.
  • Register the agent on Blocks.
  • Deploy the worker to Railway.
  • Test both local and hosted agent runs.
  • Know what prerequisites and environment variables are required.

Before You Start

  • Blocks account, so the agent can be registered and called by other agents or workflows.
  • Blocks API key, so the local and Railway-hosted provider can connect to Blocks.
  • Railway account, so the agent can run continuously online.
  • Railway CLI authentication, so the builder can create the project, set variables, and deploy.
  • Apify account and API token, so the agent can collect public profile details and post activity.
  • Apify billing or actor access, because live actor runs may incur cost.
  • LinkedIn profile URL under /in/, which is the business input the report is based on.
  • Node.js and npm, so the TypeScript agent can be installed, checked, and run.
  • GitHub access, if you want to publish or update reusable instructions like this skill file.

How It Comes Together

First, the builder creates a Blocks provider agent with a clear input: a LinkedIn person profile URL and a few options such as the post window and maximum number of posts.

Next, the agent calls Apify for two kinds of public data: profile details and recent posts. The agent normalizes the returned data because scraping tools can change field names over time.

Then the agent creates two outputs. The markdown report is for people to read. The JSON file is for debugging, auditability, or another agent that wants structured evidence.

Finally, the builder registers the agent on Blocks and deploys the worker to Railway so it can run online without keeping a local terminal open.

Starter Prompt

Use this reusable skill file to build a LinkedIn person report agent:
https://raw.githubusercontent.com/CagedEther/agentic-business-schooled/refs/heads/main/linkedin-person-report-agent-skill.md

Build a Blocks agent named [agent_name] that accepts a LinkedIn person profile URL, uses Apify for public profile details and recent activity, returns a short markdown report and structured JSON evidence, and deploys the provider to Railway. Keep comments and reactions off by default unless I ask for audience research.

What a Good Result Looks Like

A good report is short, specific, and traceable. It should name the person’s current role, summarize visible activity patterns, list a few recurring themes, cite notable recent posts when available, and include practical conversation starters.

A good technical result has a running Blocks agent, a Railway deployment in a healthy state, clear required environment variables, and a successful hosted test that returns both person_report and research_json.

Use It Responsibly

Use this only with public profile and activity data returned by the configured tools. Do not include private credentials, hidden data, or sensitive customer notes in the report. Keep comments and reactions disabled unless the business case truly requires audience-level research, because those options can increase cost and expand the amount of personal data collected.

LinkedIn activity can be incomplete. Some profiles have no recent public posts, and the best first fallback is to rerun with a wider post window such as postedLimit: "any".

What to Try Next

Add a relationship-prep mode that lets a sales or partnerships team enter the meeting goal, then tailors the conversation starters to that goal while still linking back to the public evidence.