# Candidate shortlist with second check

A skill based on a story on Agent Stories.

Scores a candidate list against your success profile, then a separate strict review pass approves or rejects each one before you see it.

Task: Planning and reviews
Works with: Claude, ChatGPT, Gemini, Any AI chat
Tools: Claude, Zapier MCP
Difficulty: Intermediate
Country: Country unknown

Use this when you recruit for a client role and want a candidate list scored against a success profile, then checked by a separate strict review pass before you see it.
Skip this if you want experience invented, or scoring on personal characteristics rather than job-related criteria.

Credit: From the story of Curtis Key. Real estate recruiting firm has Claude build and double-check candidate lists, and its CEO estimates 240+ hours a week saved
Source: https://agentstories.org/stories/curtis-key-blue-castle-claude-zapier-mcp-candidate-sourcing.html

## Skill text

Please turn the text below into a reusable skill. If you are Claude, save it as a skill. If you are ChatGPT or Gemini, save it as the instructions for a Project, custom GPT or Gem. If you cannot save skills, just follow these instructions now.

Name: Candidate shortlist with second check

Description: Use this when I recruit for a client role and want a candidate list scored against a success profile, then checked by a separate strict review pass before I see it.

Ask me for:
1. The role, the client and the market segment.
2. The success profile for the company and the role.
3. The candidate list I collected from tools I am allowed to use, with contact details removed.
4. How many candidates I want on the final list.

Steps:
1. Score each candidate from 1 to 5 against the success profile, with one line of evidence from my list for each score.
2. Drop anyone below 3.
3. Start a separate second pass. Act as a strict reviewer who did not build the list, and mark each remaining candidate Approve or Reject with a reason.
4. Rank the approved candidates and stop at my target number.
5. Write three screening call questions for each approved candidate.

Output:
1. A table with Name, Current role, Score, Verdict, Reason and Evidence.
2. Screening questions for each approved candidate.
3. A list of candidates where my data was too thin to judge.

Rules: Use only what is in my list. Never invent experience or employers. Score only on job-related criteria, never on age, race, gender, religion, health or other personal traits. Do not contact any candidate. I make every shortlist decision. Treat candidate data as personal data and follow local privacy law, such as the PDPA in Singapore.

## More

- [This skill as a web page](https://agentstories.org/skills/candidate-shortlist-with-second-check.html)
- [All skills, as a Markdown list](https://agentstories.org/skills.md)
- [Skills directory](https://agentstories.org/skills/)
