How to use these
Copy a prompt into Claude (or your AI assistant of choice), fill in the bracketed parts, run it, and iterate. These produce a strong first draft — a person reviews and approves before anything reaches a candidate or employee. People decisions especially need a human in the loop. Keep what works in your own context library.
Job description
What it’s for: a clear, honest JD that attracts the right person and screens out the wrong ones.
Draft a job description for a [role] at a [revenue] [industry] company.
Include: a plain-English summary of what they'll actually do, the 5-6 outcomes they own in year one, must-have vs. nice-to-have qualifications, and how success is measured. Avoid corporate filler and inflated titles.
Tone: direct and human. Make a strong candidate want to apply and a poor-fit candidate self-select out.Screening rubric
What it’s for: a consistent, bias-aware way to compare candidates.
Build a screening rubric for the [role] JD below:
[Paste JD]
Create 5-6 weighted criteria with a 1-5 scale and a short description of what a 1, 3, and 5 looks like for each. Add 3 disqualifiers. Keep it focused on evidence of the outcomes the role owns — not pedigree.Interview guide
What it’s for: structured questions that actually predict on-the-job success.
Generate a structured interview guide for [role]. For each of the top 4 competencies, give 2 behavioral questions and what a strong vs. weak answer sounds like. Add 3 questions that probe for [the specific risk in this hire, e.g., "will they thrive without much structure?"].
Keep it to a 45-minute conversation.Onboarding plan
What it’s for: a 30-60-90 that gets a new hire productive fast.
Build a 30-60-90 day onboarding plan for a new [role] at [company].
For each phase: the outcomes they should hit, who they should meet, what they should be reading/learning, and the one thing that signals they're on track. Make day 1 and week 1 concrete.Performance review drafting
What it’s for: turning notes into a fair, specific review.
Draft a performance review for [name], a [role], based on these notes:
[Paste notes / examples / metrics]
Structure: what's going well (with specific examples), what needs to improve (specific, actionable), and 2-3 goals for next period. Be candid but constructive. No vague praise, no surprises.Policy drafting
What it’s for: a plain-language policy people will actually read.
Draft a [policy topic, e.g., remote-work / PTO / AI-use] policy for a [size] company.
Plain language, one page. Cover the rule, the why, the exceptions, and who to ask. Avoid legalese. Flag anything I should run past employment counsel before publishing.Engagement survey analysis
What it’s for: finding the signal in survey results.
Here are our engagement survey results:
[Paste scores + open-text responses]
Synthesize: the 3 strongest themes, the 3 biggest concerns (with the verbatim language people used), any difference by team/tenure, and the two actions most likely to move the needle. Distinguish what the data shows from what's a hunch.Want this wired into your people processes?
This is a starter set. Inside an AI Office engagement, your strategist turns the highest-value of these into running workflows — intake, onboarding, and knowledge that live in your systems — with a person approving every people-facing decision. When one is worth shipping as a tool, a Value Sprint delivers it in weeks. Start a conversation at frogslayer.com.