Hire faster.Without hiring worse.
Most hiring is improvised — and it shows. The job post is a copied wish list. Interviews wander. Candidates get compared on who interviewed best, not who fits the role. Good people get passed over for the wrong reasons, and the process drags long enough to lose the ones you wanted.
The Recruiting & Hiring Skills Pack is six Claude skills covering the full hiring lifecycle — attract, screen, interview, decide, communicate, close — built to one standard: fair, structured, defensible. You hire faster and on evidence. One-time $79. Not legal advice; the skills tell you what to have HR or counsel review.
Generic AI makes hiring faster — not better.
Most growing companies hire badly and know it. Copied job posts, gut-feel interviews, slow follow-up that loses the candidates you wanted. The first instinct for most teams is to drop a JD into ChatGPT and ship faster — which gives you a slick posting full of the same coded language, interview questions you legally shouldn’t ask, and confident output with zero fairness check.
Faster bad hiring is still bad hiring. The fix is at the process layer: structured criteria, defined questions, evidence-based scorecards, and explicit flags on the language and questions that are commonly risky. That’s what actually moves the needle on fair and predictive hiring — and what generic AI doesn’t default to.
The job post lists every "nice to have" alongside the actual must-haves. Good candidates self-select out; the inflated bar reads as exclusionary; the loaded language ("young and energetic," "culture fit") screens out the people you needed most.
"So, tell me about yourself" — the improvised chat that predicts almost nothing. Different questions for different candidates makes the comparison meaningless and the decision indefensible. Structured interviews are the single most evidence-backed practice for predictive hiring.
The debrief turns into "I liked them" vs "I didn’t click." The loudest voice wins. The best small-talker beats the best fit. Scorecards tied to criteria turn impressions into evidence — and the decision into one you can stand behind.
Clear about the lane. No inflated promises.
- Six Claude skills covering the full hiring lifecycle — attract, screen, interview, decide, communicate, close.
- Three operating principles enforced at the skill level: fair by design, structured (not gut-feel), defensible (and honest about limits).
- Built on structured interviewing — the most evidence-backed practice for fair and predictive hiring.
- Honest about limits: explicitly not legal advice; the skills tell you what commonly needs counsel / HR review.
- Built on June 2026 Anthropic Agent Skills format; install once, fire automatically when the work calls for it.
- Works inside your stack — Greenhouse, Lever, Ashby, BambooHR, Notion, or no ATS at all.
- Legal advice or a compliance guarantee. Employment law varies by jurisdiction and changes — the skills flag what to check, not what's safe.
- An ATS replacement. It produces the thinking and the documents; Greenhouse / Lever / Ashby track applicants and schedule interviews.
- A recruiter replacement. Judgment, relationships, sourcing networks — still on you.
- A bias eliminator. It reduces common bias and surfaces risky territory. No tool honestly eliminates bias.
- A subscription. One-time $79 with 12 months of skill updates.
- A guarantee of great hires. Better process is the floor; the people-judgment is still yours.
One pack. The full hiring lifecycle.
Each skill handles one stage of the hiring lifecycle, and the skills compose — the JD feeds the screener; the screener feeds the interview kit; the interview kit feeds the scorecard; the scorecard feeds the offer.
“Write a job description for X,” “improve this JD,” “make this job ad more inclusive,” “check this posting for bias”
Clear, compelling JD structured around the role's real must-haves (not a wish list). Flags loaded language (age cues, gendered terms, coded “culture fit,” inflated requirement lists) and rewrites it inclusively. Ships with an inclusive-JD reference covering the patterns to hunt.
“Screen these resumes,” “build a screening rubric for this role,” “score this candidate against the role,” “shortlist candidates”
Turns a stack of résumés into a consistent, defensible shortlist. Evaluates every candidate against the same defined criteria instead of first-impression gut feel. Flags factors that shouldn't influence the decision. Ships with a screening-rubric reference.
“Build an interview kit for X,” “interview questions for this role,” “create a structured interview guide,” “what should I ask a [role] candidate”
Structured interview guide: role-based questions grouped by what they assess, scoring guidance, and an explicit flag on question territory that's commonly unlawful or risky to ask. Built on structured interviewing — the most evidence-backed practice for fair and predictive hiring. Ships with a lawful-questions reference.
“Make a scorecard for this role,” “score this interview,” “run a hiring debrief,” “compare these candidates fairly”
Scorecards tied to the role's competencies + a structured debrief that turns “I liked them” into evidence. Comparable across candidates and interviewers, so the hire is the best fit — not the best small-talker or the loudest voice in the room.
“Write a rejection email,” “reach out to this candidate,” “update the candidates who didn't hear back,” “invite someone to interview”
The messages most teams rush, delay, or skip — outreach, status updates, respectful rejections, follow-ups. Written on the principle that how you treat the people you *don't* hire is your reputation. Every candidate stays a potential future hire, referrer, or customer.
“Draft an offer letter for X,” “write a job offer,” “help me close this candidate,” “put together an offer email”
Clear, warm offer-letter draft + close plan. Explicit prompts to have your counsel or HR review before sending — because offer letters carry real legal and financial consequences. Speed on the draft, discipline on the sign-off.
Fair. Structured. Defensible — and honest about limits.
The six skills are the surface. The three principles below are the IP — pasted into every skill, enforced on every output, and the difference between AI that makes hiring faster but worse and AI that makes it faster and better.
Loaded language flagged and rewritten. Every candidate evaluated against the same criteria. The skills reduce common bias and surface risky territory — the evidence-backed levers for fairer hiring. (Honestly: they reduce bias; no tool eliminates it.)
Defined criteria, structured interviews, evidence-based scorecards. Consistency is what makes a hiring decision both better and defensible — and it's exactly what ad hoc processes lack. The most evidence-backed practice for predictive hiring.
The skills flag what commonly needs legal or HR review and tell you to check it. They are not legal advice and don't promise compliance — they make you more careful, not falsely confident. That honesty is the point.
What the bias-check actually does.
Below is the abridged job-description-writer/SKILL.md file. June 2026 Anthropic format, deployable as shown. Behavioral guardrails live in the body; the patterns it hunts for live in the references.
---
name: job-description-writer
description: Use whenever someone is writing or improving a job description, job posting, role spec, or “help wanted” ad. Triggers on “write a job description for X,” “improve this JD,” “post for a [role] opening,” “make this job ad more inclusive,” “check this posting for bias.” Produces a clear, compelling, structured JD AND runs a bias check — flags loaded, exclusionary, or commonly unlawful language (age cues, gendered terms, coded “culture fit,” inflated requirement lists) and rewrites it inclusively. Always frames bias guidance as not legal advice and reminds the user that employment law varies by jurisdiction.
---
When asked for a job description:
1. Identify the role's real must-haves vs nice-to-haves. Push back gently on inflated requirement lists.
2. Write the JD structured around must-haves, responsibilities, team context, compensation transparency where appropriate.
3. Read references/inclusive-jd-guide.md for the bias patterns to hunt: age cues, gendered terms, coded “culture fit,” inflated lists, exclusionary experience anchors.
4. Flag any matches in the draft. Rewrite each inclusively with a one-line explanation of what was loaded.
5. Append the “Have HR or counsel review” reminder. This is not legal advice and law varies by jurisdiction.
NEVER:
- Assert that a phrase is legally safe or unsafe. Flag what's commonly risky; the user verifies with counsel.
- Silently rewrite without naming what was flagged. The learning is in the explanation.
- Pad with corporate filler (“dynamic team,” “rockstar,” “ninja”). These hurt the posting and screen out serious applicants.
ALWAYS:
- Surface the bias-check output as a separate, named section the user can review before shipping the JD.
- Append the not-legal-advice reminder with the jurisdiction caveat.Six stages. Six skills. One spine.
Each skill hands off cleanly to the next. The JD produces the criteria the screener uses; the screener’s shortlist feeds the interview kit; the interview kit’s questions feed the scorecard; the scorecard’s decision feeds the comms and the offer.
→ bias-checked JD that screens no one out for the wrong reasons
→ shortlist scored against the same criteria, factors-to-ignore flagged
→ structured guide, role-based questions, lawful-territory flagged
→ evidence-based debrief, comparable across candidates and interviewers
→ respectful messages — outreach, updates, warm rejections
→ fast draft, explicit counsel/HR review prompt before sending
The integrity moat.
Exactly what you get for $79, and what you don’t.
- Six production-grade Claude skills (job-description-writer, candidate-screener, interview-kit-builder, scorecard-generator, candidate-communication, offer-letter-drafter).
- Three reference files (inclusive-jd-guide / screening-rubric / question-bank-and-lawful-questions) loaded by the matching skill.
- Three operating principles (fair / structured / defensible) enforced at the skill level.
- 12 months of skill updates as the catalog evolves.
- Setup + per-role deployment guidance (per-role Claude Project pattern).
- Honest about limits — explicit not-legal-advice framing in every skill that touches legal-adjacent territory.
- Legal advice or a compliance guarantee. Have your counsel review.
- A Claude subscription. You need Claude Pro / Max / Team / Enterprise separately.
- An ATS or applicant tracking. Greenhouse / Lever / Ashby / BambooHR still own the workflow records.
- Candidate sourcing or relationship management. Your recruiter / hiring manager still owns those.
- Background checks, reference checks, automated bias scoring.
- A subscription. One-time $79 with 12 months of updates.
Pairs naturally with the Fractional Executive & Consultant Skills Pack ($99) for fractional CHRO and people-ops engagements — that pack handles the engagement-level deliverables (diagnostic, 90-day plan, board updates); this pack handles the role-by-role hiring execution.
For agency teams hiring on behalf of clients, pair with the Agency Operators Skills Pack ($89) — agency client comms + scope discipline on top of consistent per-client hiring process.
For founders running first hires solo, pair with the Solo Founder Skills Pack ($79) — investor updates, board memos, and weekly cadence around the new hire calibrated for the same operator.
The questions founders and people leads actually ask before deploying.
Six Claude skills covering the hiring lifecycle: a bias-checked job-description writer, a structured candidate screener, an interview-kit builder, a scorecard generator, a candidate-communication writer, and an offer-letter drafter. Each installs as a SKILL.md folder that Claude loads automatically when the work calls for it, and each is built to make hiring more fair, more consistent, and more defensible — not just faster.
Founders and managers who hire without a dedicated recruiting team; operations and people/HR leads building a repeatable hiring process; and recruiters or agencies who want consistency across roles and clients. If you're hiring and the process is ad hoc, inconsistent, or slower than it should be, this is built for you.
No. These skills are not legal advice and do not guarantee compliance. Employment law varies by country, state, and city and changes over time. The skills are designed to reduce common bias and flag language and interview questions that are frequently risky — but they explicitly tell you to have your own counsel or HR professional review anything before you rely on it. Treat the flags as prompts to check, not as legal clearance.
Three ways. The job-description writer flags loaded language (age cues like “young and energetic,” coded “culture fit” phrasing, gendered terms) and rewrites it inclusively. The screener and interview kit push you toward evaluating every candidate against the same defined criteria instead of gut feel, which is the single most evidence-backed way to make hiring fairer. And the interview kit flags question territory that's commonly unlawful or risky to ask. It reduces bias and surfaces issues; it does not claim to eliminate bias — no tool honestly can.
A prompt pack is text you paste and re-paste. A skill installs once into Claude and fires automatically when the task matches — you ask for an interview kit and the interview-kit skill runs, to the same standard, every time. The skills also encode a hiring standard (structure, fairness checks, guardrails) rather than a one-off instruction, so every role you hire for follows the same defensible process.
Each skill is a folder with a SKILL.md file. Add them to Claude.ai (Settings → Capabilities → Skills) or to your Claude Code skills directory. Once installed, they trigger automatically when your request matches. They also work dropped into a Claude Project — a clean setup if you keep a Project per open role.
No. This produces the thinking and the documents — job descriptions, screening rubrics, interview guides, scorecards, candidate messages, offer drafts. It doesn't track applicants, schedule interviews, or replace a recruiter's judgment and relationships. It makes whoever is hiring faster and more consistent, inside the tools and process you already use.
Different operating layer. Fractional Executive handles engagement-based deliverables for advisors selling judgment (diagnostics, proposals, board updates). Agency Operators handles client-facing agency process (onboarding, scope creep, status reports, renewals). This pack is the function — hiring — calibrated for the role, not the practice. Most operators end up running two of these packs together; they compose without overlap.
30-day no-questions refund. Run the job-description writer and the interview-kit builder on one real open role this week. If they don't give you something more fair, more structured, and faster than your current process, email and we refund.
Hire faster.
Without hiring worse.
Six Claude skills covering the full hiring lifecycle. Three operating principles enforced at every output. Honest about limits, including what to have counsel review. One-time $79.
Sold by RedHub AI LLC · Secured by Stripe · redhub.ai