audience-filterlisted

Use when stage 6 of the prospecting system has a raw audience — page followers, post engagers, community members, event attendees, or discovered companies — and the ICP judge prompt has to be built, tuned with a human, and applied before any money is spent on enrichment. Also use when a list came back with obvious bad fits.
automatewithuday/gtm-prospecting-system · ★ 1 · AI & Automation · score 75

Install: claude install-skill automatewithuday/gtm-prospecting-system

# ICP Prompt Builder ## Pipeline contract **Stage 6 of 20.** Reads `05-audience-raw/*.csv` and `02-client-profile.yaml`. Writes `06-judge-prompt.txt` and `06-audience-qualified.csv`. - Normalize every raw CSV to the lead columns in [`/prospecting-system`](../prospecting-system/SKILL.md) first. Keep `source_type` (follower / engager / member / attendee) and `source_detail` (which page) on every row. - Dedupe by `linkedin_url`, else `domain+first_name+last_name`, BEFORE judging. A person following three competitor pages is one row with three sources — that overlap is a signal, keep the count in `source_count`. - Drop rows whose company is in `excluded_domains`. - The judge sees people here, not only companies: title + company description. Same 10-row approval loop. - Output adds `qualified` (true/false), `judge_reason`. Rejected rows stay in the file. Files live in `runs/<run>/`. Column definitions: [`/prospecting-system`](../prospecting-system/SKILL.md). Standalone use: ignore this block. Before you pay to pull 5,000 companies, tune a qualification prompt on 10-50 of them. This skill walks you through the iterative loop. ## Why this exists List-builder skills (DiscoLike, Blitz, Prospeo, Google Maps) return COMPANIES, but they don't know whether those companies match your ICP. If your list-builder returns 5,000 companies and 80% are wrong fits, you'll waste money enriching them for emails that go nowhere. The fix: build an AI qualification prompt BEFORE scaling. Pull 1