Candidate Scorecard Builder
Without a scorecard, hiring decisions collapse into 'I liked them' — which is mostly affinity bias wearing a professional label. The Candidate Scorecard Builder produces a weighted rubric with behavioural anchors at every score, so two interviewers can rate the same candidate and end up within a point of each other.
What this skill does
The scorecard isn't bureaucracy — it's the difference between hiring the best candidate and the most charming one. Most hiring decisions collapse into "they seemed strong" or "I had a good feeling," which are clean restatements of three biases: affinity (they remind me of me), halo (one great answer coloured the whole evaluation), and recency (the last candidate had an unfair advantage). A scorecard doesn't remove judgment. It structures it — same criteria, same scale, same weight, every candidate.
The framework runs three tiers — Critical, Important, Desired — split 60/30/10 across 100 points. Critical competencies are dealbreakers; a score of 1 on any of them is an automatic No Hire regardless of total. Important competencies differentiate; Desired competencies break ties. Weights are set by job impact, not by how testable a competency is. Technical skill is often easiest to assess — but if collaboration is what actually determines success in the role, collaboration gets the weight.
Every competency gets behavioural anchors at every level. Without them, "3" means whatever the interviewer wants it to mean, and the scale dissolves. The 4-point scale is deliberate — five points lets people drop a "3" on everything and avoid a decision. Four forces a call: above the bar, or below it.
The skill also produces a stage-aware version. A phone screen scores 2–3 must-haves, not the full sheet. A technical round weights technical competencies at 70%+ of the stage score. A final round runs the full scorecard. Using the same detailed form at every stage produces scorer fatigue and worse data.
Two structural rules sit underneath the output. Scorecards are completed independently before any group discussion — otherwise the highest-status person anchors the room. And evidence supports every score: "they described a specific example of X" is a score; "I had a feeling" is not. The comparison view then lays candidates side by side, weighted totals visible, so the decision is grounded in what the panel actually saw — not who told the best story in the debrief.
When this triggers
- ·You're hiring for a role and the interview panel has wildly different opinions after every debrief
- ·You keep hiring people who interview well and underperform — your filter isn't measuring the right things
- ·You need to defend a hire/no-hire decision to a co-founder, board, or unsuccessful candidate
- ·Your last three hires looked nothing like each other on paper and the team is now uneven on key competencies
- ·You want a structured way to compare internal and external candidates without the familiarity discount or halo
Example
Trigger
User: 'Hiring a Senior Software Engineer. Two interviewers, three rounds, never agree on the final candidate. Build me a scorecard that ends the arguments.'
Output
Weighted competency framework (100 points): · Technical depth — Critical — 25 · System design ability — Critical — 20 · Problem-solving approach — Critical — 15 · Collaboration — Important — 15 · Communication clarity — Important — 15 · Mentoring capability — Desired — 5 · Domain experience — Desired — 5 4-point scale (no lazy middle): 4 Exceptional · 3 Strong · 2 Developing · 1 Does Not Meet · N/A Behavioural anchors per competency. Example — System design: · 4: Articulated trade-offs unprompted, proposed approach the interviewer hadn't considered · 3: Solid design, identified key trade-offs when asked · 2: Basic design, missed significant trade-offs · 1: Could not articulate coherent design Decision rules: · Score of 1 on any Critical = No Hire (regardless of total) · Min 3 on all Critical competencies for Hire threshold · Scorecards completed INDEPENDENTLY before debrief Plus: stage-specific weighting, comparison view across candidates, calibration exercise for new interviewer cohorts.
Get this skill + 7 more
Included in the The Startup Founder Stack — move fast with ai leverage. Save $150+ vs buying individually.
Get The Startup Founder Stack — $149What you get
- 239-line SKILL.md, ready to drop into ~/.claude/skills/
- Tested through 3 Karpathy-loop iterations (versions v1.0.0 → v1.3.0)
- Triggers automatically when relevant — no command to remember
- Lifetime updates as the skill is refined further
More from Recruitment
Builds questions that distinguish between culture FIT (shared values and working principles) and culture CLONE (same background, same personality, same demographics)
Doesn't produce generic questions — it produces questions that actually predict job performance because they're tied to the real capabilities the role requires
Writes job descriptions that sell the opportunity while being honest about the role. Activelies remove language that discourages qualified candidates from applying (especially gendered language,…
Produces offer letters that are legally sound, warm enough to reinforce the candidate's decision, and specific enough to prevent misunderstandings
Doesn't produce a first-day orientation checklist — it builds a three-month ramp-up plan that transforms a new hire from "person who just started" into…
Produces pitches that lead with what the candidate cares about (career growth, impact, culture, compensation), not what the company wants (a warm body to fill a seat)
Browse the full library
297 skills across 31 categories. One purchase, lifetime updates.
See all bundles