Pre-send list clearance skill

What it does

1. Reads your CSV and finds the email column.
2. Trims and lowercases addresses, removes duplicates, and flags invalid syntax, role addresses and known disposable domains.
3. Verifies every other address with the SimpleVerifier API, 100 per request.
4. Sorts each row by the API's send_advice value and writes three files.
5. Prints a summary: rows read, duplicates, and rows per file with a count for each reason.

Every output row keeps your columns and adds sv_status, sv_send_advice and sv_reason.

Install

Claude Code

unzip simpleverifier-list-clearance-skill.zip -d ~/.claude/skills
export SIMPLEVERIFIER_API_KEY="your key"
claude

Use .claude/skills inside a project to share it with your team. Then ask: “Clear leads.csv before I send it.”

Codex and Cursor

Unzip the folder into the skills folder your version of the tool reads. Folder names differ between versions, so check the tool's own docs. If your version has no skills folder, put the unzipped folder in your project and tell the agent: “Run the list-clearance script on leads.csv.” The script runs on plain Node 18 or newer.

Run it without an agent

export SIMPLEVERIFIER_API_KEY="your key"
node list-clearance/scripts/list-clearance.mjs leads.csv --out clearance-out

Add --column NAME if the email column has an unusual header, or --local-only to run the local checks with no key. The key is read from the environment and never written to disk.

Use it in a Clay workflow

1. Export the Clay table as a CSV.
2. Run the skill on the export from Claude Code, Codex or Cursor.
3. Import clean.csv back into Clay, or straight into your sequencer. Keep hold.csv out of cold campaigns.
4. Add suppress.csv to your suppression list.

Want a verification column inside the table instead? The Clay integration shows the HTTP API setup.

Cost and limits

The skill is free. The verification runs on your SimpleVerifier plan through the API, and API calls are metered past the amount your plan includes. For flat-price verification of very large lists, upload the CSV on the verify page. Agents that call the API or MCP all day: see verification for AI agents, or the MCP server docs.

Source: integrations/agent-skills/list-clearance. By the SimpleVerifier Team.

Common questions

What is the list clearance skill?

A free Agent Skill: a folder with a SKILL.md file and one Node script. Given a CSV of leads, it removes duplicates, flags bad syntax, role and disposable addresses, verifies the rest with the SimpleVerifier API and writes clean.csv, hold.csv and suppress.csv with a reason column.

Does it work in Claude Code, Codex and Cursor?

The skill is a standard SKILL.md folder, and the script runs on plain Node 18 or newer, so any agent that can read a skill folder and run a command can use it. Claude Code reads skills from .claude/skills. For Codex and Cursor, copy the folder into the skills folder your version uses, or ask the agent to run the script directly.

Do I need a SimpleVerifier account?

You need an API key, which comes with the Pro, Business and Agency plans. Create it on the API keys page and set it as SIMPLEVERIFIER_API_KEY. The skill never stores the key. Without a key you can still run the local checks with --local-only.

What does it cost to run?

Pro is $39 a month. API calls are metered: Pro includes 10,000 calls a month, then $1.50 per 1,000. The script sends up to 100 addresses per request. For very large lists, a CSV upload on the verify page is the route for big lists: the first 100,000 checks a month at full speed, then slower, never cut off.

Which send_advice values does it use?

safe goes to clean.csv. care and retry go to hold.csv. do_not_send goes to suppress.csv. A role address stays in hold.csv even when the mailbox is confirmed.

Can I use it in a Clay workflow?

Yes. Export the Clay table as a CSV, run the skill, then import clean.csv back into Clay or into your sequencer. Clay's own Agent Plugin can load skills too. If you want a verification column inside the table instead, the Clay integration page shows the HTTP API setup.

What are the limits?

The API allows 120 requests a minute per key, and the script waits when told to slow down. A verifier reports what the mail server says: catch-all and hidden-provider addresses come back as care, never as safe.