Social Lead Researcher — AI skill for Claude Code

Social Lead Researcher

**Give your AI coding assistant a job: find qualified prospects on LinkedIn and X — and don't stop until it's done.**

Social Lead Researcher is a tiny, open, do-it-yourself social selling research worker. You install it once into [Claude Code](https://claude.com/claude-code), [OpenCode](https://opencode.ai), or [Codex CLI](https://github.com/openai/codex). From then on, you just say:

*"Find 500 qualified prospects for me on LinkedIn and X based on my ICP, and don't stop until it's done."*

…and your AI goes to work: researching real people from **public** profiles and posts, checking each one against **your** Ideal Customer Profile, capturing the recent public signal that makes them hot, recording only the ones that pass the quality gate — batch after batch, session after session — until the target number is reached. Progress is saved to disk: close your laptop, reopen, say "continue", and it picks up where it left off.

**Research only. It never sends messages or connection requests — you decide who to contact.** No accounts. No servers. No subscriptions. Your prospect data never leaves your computer.

---

What you get

🎯 A real target	The goal (500) is stored in data/state.json — not in chat memory. It survives restarts.
✅ A quality gate	A prospect only counts if score ≥ 70 and evidence cites your ICP file and there's a real profile URL + recent public signal. No junk padding.
🔥 Hot-not-cold	Every record carries a recent public signal (post, comment, launch, hiring) — so your outreach list is people who are active right now.
🛡️ Human decides contact	The tool researches; it never messages anyone. Ever.
🔁 Continuous work	Batch loop with resume across sessions.
📝 One editable ICP	Everything about who counts lives in one human-readable file: icp.md.
📤 Clean export	python3 sr.py export csv → prospects.csv, sorted by score, Link