Source developers through the open-source work they publish

Developers who maintain or contribute to projects in your stack, their GitHub activity this year, and a work email for the shortlist.

Steps

At most, per call

Open-source sourcing finds engineers through the code they publish, not their job titles. With Agentik, your AI finds active projects on GitHub, reads the projects, contributors and their activity through ScrapeCreators, and finds work emails with Icypeas.

Step 1: Find active projects
GitHub search finds active repositories in your stack, ranked by stars, and ScrapeCreators reads the five best projects.
- GitHub repository searchGitHub · 2 searchesFree
- GitHub repositoryScrapeCreators · 5 projects$0.0094
Step 2: Read the contributors
ScrapeCreators reads fifteen contributors’ profiles and their contributions this year.
- GitHub userScrapeCreators · 15 developers$0.03
- GitHub contributionsScrapeCreators · 15 developers$0.03
Step 3: Find emails
Icypeas finds work emails for the ten best.
- Icypeas email finder in batchIcypeas · 10 people$0.19
Step 4: Write the messages
Your AI writes the shortlist and a message for each that mentions their work.
Your agent does this part, with no tool call.

Cost of the example

At most $0.26one run

2 repository searches, 5 projects read, 15 developers’ profiles and contributions, 10 emails found.

Give this to your agent

Use my connected Agentik MCP server for this job: developer sourcing from github.

Source developers from open source. First ask me for the technologies, the kind of projects and the location if it matters.

Find active projects with github/search-repositories and read the five best with scrapecreators/get-v1-github-repository. Read 15 contributors’ profiles with scrapecreators/get-v1-github-user and their contributions this year with scrapecreators/get-v1-github-user-contributions. Find work emails for the ten best with icypeas/find-emails-batch.

Finish with a shortlist (developer, projects, activity, location, email) and a message for each that mentions their work.

Before running anything, inspect each tool with inspect_tool, ask me for any input you don’t have instead of inventing it, and tell me the total price. Wait for my go. At the end, tell me what Agentik charged.

3 providers, one connection

What you get

A developer shortlist with emails.

You provide

The technologies
The kind of projects
Location, if it matters

Your agent returns

Developers active in your stack
Their projects and activity this year
Work emails
A message that mentions their work

DevelopersExample layout

Developer	Projects	Location	Email
Developer A	Vector database core	Berlin	Found
Developer B	Client library	Remote	Found
Developer C	Benchmarks	Unknown	Not found

Placeholders. Your rows come from your run.

Each provider does one part

No single provider covers this job. Your agent calls each one for its part, from one connection.

Good to know

GitHub activity shows public work only; many great engineers work in private repositories.
Reach out about their work and respect people who say no.

When to use something else

For a database of every developer’s skills, use a technical sourcing platform. This playbook finds people through the work they publish.

Questions

What about engineers with private work?

They are invisible here. Public activity shows public work only.

How should I reach out?

About their work, briefly, and respect anyone who says no.

What does the example cost?

At most $0.26, reserved before each call and settled at the measured usage.

Connect your agent to 2,400+ tools for free

Get $5 free credit Get $5 free credit