# Buyer Intent Post Finder

> Find recent LinkedIn posts where someone is actively asking to buy the service you sell. Use when the user wants fresh B2B leads from LinkedIn posts, wants to build a search query list, or asks "find people who need <service>".

- Skill: `syp1xd/buyer-intent-post-finder` (Agent Skill)
- Install (CLI): `npx skillmds@latest add syp1xd/buyer-intent-post-finder`
- Raw SKILL.md: https://api.skillmd.com/api/skills/syp1xd/buyer-intent-post-finder/raw
- Safety review: pending
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: Coding & Dev Tools
- Author: syp1xd (https://skillmd.com/u/syp1xd)
- Updated: 2026-09-22
- Page: https://skillmd.com/skills/syp1xd/buyer-intent-post-finder

---


# Buyer-intent post finder

Goal: a short list of posts written by BUYERS, not vendors, from the last 14 days.

## 1. Build the query list
Ask the user what they sell. Then write queries in three tiers and save them to `queries.txt` (one per line, `#` for comments):

- **Tier 1 - words only a buyer writes.** Buyers ask for the artifact or a qualified vendor: "looking for a <certified> agency", "<deliverable> certificate required", "need a quotation for <service>". Vendors write "we are <certified>" instead.
- **Tier 2 - explicit vendor requests.** "<service> partner", "<service> requirement", "<service> vendor", "RFP <service>".
- **Tier 3 - hiring as a demand signal.** A company hiring its own person for the work often outsources part of it: "<role> hiring".

## 2. Run the searches
- Run each query **verbatim, one per search**. Do not merge queries with OR and do not add loose words. LinkedIn ANDs extra words in, so a merged query returns fewer posts, not more.
- Filter: Posts, Latest, Past week.
- Work Tier 1 first. It has the highest yield.

## 3. Filter
- Drop posts older than 14 days or with an unknown date. Older requirements are usually awarded already.
- Read the complete post before you keep it.
- Dedupe on the author's profile URL. One good post matches many queries.
- Record for each: author, profile URL, company, post URL, post date, every query that matched, and a snippet long enough to judge the post without opening LinkedIn.

Pass the list to `hot-warm-lead-classifier`.