AI Prompts Every Recruiter Should Know

Most AI advice for recruiters stops at "use ChatGPT to write job descriptions." That is about as helpful as telling someone to "use Excel for data." The tool matters less than what you ask it to do, and how you ask.

The difference between a useless AI output and a useful one almost always comes down to the prompt. A vague prompt produces vague output. A specific prompt with the right context and constraints produces something you can actually use.

Here are prompts for the tasks that eat the most recruiter time, based on the areas where AI adoption is already highest. Industry surveys consistently rank job description writing as the most common AI use case among agencies, followed by CV screening and talent acquisition. These prompts reflect what is actually working.

Prompt 1: Job Description First Draft

This is the most popular AI use case in recruitment for good reason. Writing job descriptions from scratch is slow. Writing them from a prompt-generated first draft is much faster.

"Write a job description for a [Job Title] at a [Company Type] based in [Location]. The role reports to [Reporting Line] and the salary range is [Range]. Key responsibilities include [list 3 to 5 main duties]. Required experience: [list essentials]. The company culture is [brief description]. Write in a professional but approachable tone. Include a section on benefits if I provide them. Do not use superlatives like 'amazing opportunity' or 'world-class team.' Keep it under 600 words."

The critical parts are the constraints. Without "do not use superlatives" and "keep it under 600 words," you will get a bloated, cliche-heavy description every time. The more specific your constraints, the less editing you need to do afterwards.

Prompt 2: CV Screening Criteria

Rather than asking AI to screen individual CVs (which has compliance implications under ICO guidance), use it to build structured screening criteria that you then apply consistently.

"I am recruiting for a [Job Title] role. The essential requirements are [list requirements]. The desirable requirements are [list nice-to-haves]. Create a scoring rubric I can use to screen CVs consistently. For each requirement, define what a strong match, partial match, and no match looks like. Include specific things to look for on a CV for each criterion. Flag any requirements that might be difficult to assess from a CV alone."

This prompt does not automate screening. It standardises it. The output gives you a framework that every recruiter on the team can apply the same way, which addresses both consistency and compliance concerns.

Prompt 3: Candidate Outreach Messages

Generic outreach gets ignored. Personalised outreach takes time. AI can help bridge the gap, but only if you give it enough context about both the role and the candidate.

"Write a LinkedIn outreach message to a [candidate's current role] at [their current company] about a [your role title] opportunity. The candidate's relevant experience includes [specific details from their profile]. The key selling points of this role are [list 2 to 3 genuine advantages]. Keep the message under 150 words. Use a conversational, direct tone. Do not use phrases like 'exciting opportunity' or 'I came across your profile.' End with a specific question, not a generic 'would you be open to a chat.'"

The inclusion of specific candidate details is what separates useful AI outreach from spam. You still need to read the candidate's profile. The AI saves you 10 minutes of writing time, not 10 minutes of research time.

Prompt 4: Boolean Search String Building

Boolean search remains essential for sourcing, but building effective strings is time-consuming and error-prone. AI can generate starting points that you then refine.

"Build a boolean search string for LinkedIn Recruiter to find [Job Title] candidates. The target profile has experience in [key skills or technologies]. They should be based in [location or region]. Exclude [specific exclusions, e.g., recruiters, consultants]. Include variations and synonyms for job titles (e.g., if searching for developers, include software engineer, programmer, etc.). Format the string so I can paste it directly into LinkedIn Recruiter's search bar."

This is a starting point, not a finished product. The AI will generate a reasonable first attempt, but you will need to test and refine the string based on the results it produces. The time saving comes from not starting with a blank search bar.

Prompt 5: Interview Preparation Brief

Recruiters preparing candidates for interviews spend significant time compiling company research and likely questions. AI can accelerate this.

"Create an interview preparation brief for a candidate interviewing for a [Job Title] at [Company Name]. Research the company and include: a 3-sentence company overview, the likely interview format for this type of role, 5 likely interview questions with brief guidance on strong answers, and 3 questions the candidate should ask. Base the interview questions on what is standard for [industry/role type] interviews, not generic questions. Keep the total brief under 400 words."

The output will need checking against what you know about the specific client's interview process. But it provides a solid starting framework that is faster to edit than to write from scratch.

Prompt 6: Client Reporting Summary

Weekly or monthly client reports follow predictable formats but take time to compile. AI can draft the narrative sections from your data.

"I need to write a weekly recruitment progress report for a client. Here are the key metrics this week: [applications received], [candidates screened], [interviews arranged], [offers made], [positions filled]. Compared to last week: [brief comparison]. Write a 200-word summary that highlights progress, flags any concerns, and recommends next steps. Use a professional, factual tone. Do not pad the report with unnecessary commentary."

The constraint "do not pad" is essential. Without it, AI will produce fluffy progress reports that waste your client's time. You want concise, specific, actionable summaries.

What Prompts Cannot Do

These prompts are starting points for specific tasks. They save time on first drafts, structure, and repetitive writing. They do not replace recruiter judgment.

The decision about whether a candidate is right for a role, whether a client's expectations are realistic, or whether an outreach message hits the right tone still requires a human who understands the context. AI is a drafting tool, not a decision-making tool.

Prompts are also the entry point for working with AI, not the ceiling. The agencies seeing the biggest returns are the ones that have moved beyond ad-hoc prompting to structured AI workflows built around their specific processes. A prompt you use 50 times a week is a workflow waiting to be built.