producthunt-company-qualifier
riddhimaaan/producthunt-company-qualifier
Claude Code skill: score and label Product Hunt companies against a B2B SaaS GTM/outreach ICP
Instalação
npx convez-ai riddhimaaan/producthunt-company-qualifier1 comando no terminal: detecta o formato do repositório (marketplace ou plugin avulso) e instala via Claude Code. Requer Node 18+ e o Claude Code instalado.
Métricas
README
A tradução para o português deste README ainda está na fila.
ProductHunt Company Qualifier — Claude Code Skill
Score and label companies from any Product Hunt CSV against your own B2B ICP. Fetches live websites, runs keyword pattern scoring, and produces a ranked shortlist + split CSVs + an Excel workbook ready for Google Sheets.
Output labels: priority_now · good_fit · maybe · skip
Prerequisites
Before installing, make sure you have these three things:
| Requirement | Check |
|---|---|
| Claude Code installed | Run claude --version in your terminal |
| Python 3.9 or newer | Run python --version or python3 --version |
| Git | Run git --version |
Installation (do this once)
Mac / Linux — open Terminal and run these two commands:
Step 1 — Download the skill:
git clone https://github.com/riddhimaaan/producthunt-company-qualifier \
~/.claude/skills/producthunt-company-qualifier
Step 2 — Install Python dependencies:
pip install -r ~/.claude/skills/producthunt-company-qualifier/requirements.txt
Windows — open PowerShell and run these two commands:
Step 1 — Download the skill:
git clone https://github.com/riddhimaaan/producthunt-company-qualifier "$env:USERPROFILE\.claude\skills\producthunt-company-qualifier"
Step 2 — Install Python dependencies:
pip install -r "$env:USERPROFILE\.claude\skills\producthunt-company-qualifier\requirements.txt"
Note: If
pipgives an error, trypip3instead.
First-time setup (happens automatically)
The first time you run the skill, Claude will walk you through a one-time setup to personalize the scoring for your ICP. It will ask you:
- Which companies have you worked with or sold to? (your Tier 1 / Tier 2 clients)
- What are the strongest signals that a company is a great fit for you?
- What are adjacent or secondary signals?
- How would you describe the four outcome buckets for your leads?
Claude saves your answers and the skill is now tuned to your ICP — not anyone else's.
You only do this once. After that it goes straight to processing.
How to use it
Step 1 — Get your CSV ready
Your CSV needs these four columns (Product Hunt exports work out of the box):
| Column | Description | Example |
|---|---|---|
url | Company website URL | https://clay.com |
name | Company name | Clay |
domain | Domain only | clay.com |
tagline | One-line description | Enrich data and automate outreach |
Any extra columns you have will be kept in the output unchanged.
Step 2 — Open Claude Code
Open your terminal and start Claude Code in any folder:
claude
Step 3 — Run the skill
Type this command:
/producthunt-company-qualifier
Claude will ask for the path to your CSV file. Paste the full path, for example:
- Mac/Linux:
/Users/yourname/Downloads/producthunt-export.csv - Windows:
C:\Users\yourname\Downloads\producthunt-export.csv
Step 4 — Wait for it to finish
The script fetches each company's live website in parallel (up to 24 at a time). A typical run of 500 companies takes about 1–2 minutes.
Step 5 — Find your output files
All output files are saved in an outputs/ folder inside whichever directory you ran Claude Code from:
outputs/
├── qualified-products-broad.csv ← Every company, scored
├── qualified-products-broad-shortlist.csv ← Top 121 by score
├── shortlist-review.txt ← Human-readable preview
├── qualified-products-curated.csv ← Shortlist with your labels
├── qualified-products-priority_now.csv ← Direct ICP fit
├── qualified-products-good_fit.csv ← Strong adjacent fit
├── qualified-products-maybe.csv ← Weaker overlap, worth a look
├── qualified-products-skip.csv ← Off-ICP, disqualified
└── qualified-products-google-upload.xlsx ← Excel workbook, all tabs
The file to start with: open qualified-products-curated.csv — it has the shortlist ranked and split by your four labels.
Optional: Push results to Google Sheets
After the run, tell Claude: "push the results to Google Sheets" — it will use Composio to create a spreadsheet with separate tabs for each label.
Customizing your ICP later
Your scoring profile is saved at:
- Mac/Linux:
~/.claude/skills/producthunt-company-qualifier/references/scoring-config.json - Windows:
%USERPROFILE%\.claude\skills\producthunt-company-qualifier\references\scoring-config.json
You can edit this JSON file at any time to add new companies you've worked with, adjust scoring weights, or update your heuristic keywords. The human-readable version is in references/target-profile.md in the same folder.
Troubleshooting
"command not found: claude" Claude Code isn't installed or isn't in your PATH. Download it here.
"No module named 'requests'" The Python dependencies weren't installed. Re-run Step 2 from the installation section above.
"Missing required columns: domain, name, tagline, url" Your CSV is missing one of the four required columns. Check the column names match exactly (lowercase, no spaces).
The skill doesn't appear when I type /producthunt-company-qualifier
The skill folder might be in the wrong location. Verify the folder exists at:
- Mac/Linux:
~/.claude/skills/producthunt-company-qualifier/SKILL.md - Windows:
%USERPROFILE%\.claude\skills\producthunt-company-qualifier\SKILL.md
Requirements
- Claude Code CLI
- Python 3.9+
requests,beautifulsoup4,openpyxl(installed via Step 2 above)- Git
License
MIT