The Hiring Desk
Skills · Subagent · Command · PDF
.claude/the-hiring-desk/zip
skills/06
├ candidate-comparison
├ exit-plan
├ interview-guide
├ offer-memo
├ reference-call
├ role-scorecard
agents/01
├ cv-reader
commands/01
├ /scorecard
FIELD-GUIDE.pdf9pp
English
One payment · No account · No subscription
- ✓ Download link in your inbox in seconds
- ✓ Secure card checkout · VAT handled
- ✓ 14-day money back, no questions
- ✓ Every future revision included
- ✓ Runs in Claude Code and claude.ai
Built for Claude first, our recommended agent. Skills follow the open Agent Skills standard (SKILL.md), so they also run in any agent that reads it; subagents and slash commands are Claude Code formats.
The Hiring Desk
Score people against the standard, never against each other
six skills·one subagent·one command·9-page guide
Your agent writes the scorecard the role is really being filled against, builds the interview loop and its anchored scales, and holds every candidate to the same written standard.
Executive summary
- The problem
- Without a written standard, every interviewer leaves the room with a confident story about the candidate, and the debrief ranks people against each other instead of against the role. The hire feels well judged, so nobody asks what evidence it rested on.
- What it does
- Six skills and a read-only cv-reader write the scorecard first, give each interviewer one competency and an anchored scale, place every piece of evidence on the Ladder of Evidence, and keep impressions out of the score.
- Why take it
- A wrong hire costs a year and a team's trust. This makes every offer a decision you can explain, with the evidence and the bets written down before you sign.
For hiring managers and founders who interview without a written procedure.
In the first week you will
- 01Run /scorecard on the role you are about to open and get the mission, the dated outcomes and the substitution test result
- 02Build the loop with interview-guide: one competency per interviewer and a five-point scale with written anchors
- 03Hand the applications to the cv-reader subagent and get the evidence and the gaps per outcome, with no ranking
Contents of the Kit
- 01Six installable agent skills: role-scorecard, interview-guide, reference-call, candidate-comparison, offer-memo, exit-plan
- 02The cv-reader subagent: reads applications against the scorecard and reports evidence and gaps per outcome. Read-only, never ranks, never infers a personal characteristic
- 03A /scorecard slash command that opens a role scorecard from one sentence
- 04The Ladder of Evidence as a scoring rule: claim, account, artefact, corroboration, with the rungs a non-negotiable requires and the rungs that may never be scored
- 05Blank templates for the scorecard, the interview evidence form and the reference call record
- 06A 9-page Field Guide on hiring as a decision procedure, and plugin manifest included
One skill, as it ships
SKILL.md
role-scorecard
- Use when
- Use when a role is about to be opened, when a job advertisement is being written, when a hiring manager says "we need another person in support" or "I want to hire a head of X", or when an interview loop has to be designed and nothing is written down yet. Runs the substitution test and refuses to hand over a job description dressed as a scorecard.
- Requires
- The role, in the manager's words · The problem the role exists to solve
- Will not
- No personal or protected characteristic is ever recorded, requested or implied · No invented numbers · No legal opinion
- Returns
- Scorecard: {role title}
One of six skills in the kit · All of them install the same way
The contents, in full · six skills, one subagent, one command
- 01role-scorecardThe role written as a result rather than a person: mission in one sentence, four to six outcomes each with a measure and a date, the competencies those outcomes require, and the non-negotiables. It runs the substitution test before it hands anything over, and asks what would make the role a mistake to fill at all.
- 02interview-guideThe loop, built from one scorecard: which competency each interviewer owns, the past-behaviour questions in fixed wording, the probe sequence, and a five-point scale with written anchors at one, three and five. It emits the evidence form each interviewer files before the debrief, and refuses a scale whose anchors you have not supplied or approved.
- 03reference-callBuilt from the gaps, not from a template: the loop's open outcomes turned into questions that ask for behaviour and dates rather than opinions, with what the referee would not say recorded at the same weight as what they did. It carries the consent rule, and keeps the list of questions that are never asked.
- 04candidate-comparisonOne evidence table per candidate against the same scorecard: outcome, evidence, its rung on the Ladder of Evidence, and the gaps, with impressions recorded in a separate block the score cannot see. It produces no ranking, no shortlist and no recommendation, only what is proven, what is a gap and what would be a bet.
- 05offer-memoThe decision written down by a named person: level and terms from your figures, what was proven and at which rung, the bets being taken, the conditions, and the walk-away point set before the negotiation starts. The second half is the ninety days the offer commits to, with the outcomes as thirty, sixty and ninety-day milestones and the day-ninety review booked before day one.
- 06exit-planThe structured route out when the role is not working, assembled while nobody is angry: which of the three causes applies, the evidence with dates, the expectation in writing with a review date, the two outcomes stated in advance, the handover, and what the team will be told. It writes no termination letter and marks every step that touches employment law as a question for counsel in your jurisdiction.
- 07cv-reader (subagent)Reads a stack of applications against one scorecard and reports, per candidate, the evidence it found for each outcome with its rung, and the outcomes the document is silent on. It never ranks, scores or recommends, ignores name, photograph, address, age, nationality, gender, marital or family status and health, and says in its report that it ignored them.
- 08/scorecardOpens a role scorecard from one sentence about the role you want to fill. It never estimates a target, rejects virtues offered in place of competencies, and reports that the role is not ready to open when the manager cannot say what it must produce.
The Hiring Manual opens on an unkind finding: an unstructured interview can subtract information rather than add it, because an interviewer builds a coherent story out of almost any answer. What survives the evidence is structure, and structure is almost entirely clerical: the same questions in the same words, scales written before anyone is met, files closed before the debrief. The Hiring Desk is that clerical half done properly. No tool here ranks people, produces a hiring decision, infers a personal characteristic or gives a legal opinion.
Pairs with
The Hiring Manual is the argument these tools run: why structure beats the conversation, what a scorecard is, and why the first ninety days are the last phase of hiring rather than a separate programme. Think Like a CEO covers the seat the new person will sit under, from delegation to the accountability ladder. The Executive Desk’s first-120-days builds the same event from the arriving leader’s side, so a manager can run the two plans against each other and see where the mandates disagree before the new person does. The Allocation Desk asks the question this kit asks at the top of every scorecard: whether the role should be filled at all.
Before you buy
01Do I need Claude Code?
No. Every skill uploads to claude.ai as a zip. Subagents and slash commands are the Claude Code extras.
02I am not a programmer.
Install is copying folders into a .claude directory. The Field Guide in each kit assumes nothing beyond that.
03Will it work in my agent?
Skills follow the open Agent Skills standard (SKILL.md), so any agent that reads it can run them. Built for Claude first, our recommended agent.
04Do the tools invent numbers?
Never. Every figure is supplied by you, confirmed by you, or labeled as an assumption in the output.
05How does it arrive?
Instantly. Pay by card and the download link is in your inbox within seconds, and on the confirmation screen.
06Is this a subscription?
No. One payment, lifetime access to your download, every future revision of what you bought included.
07What if it is not useful to me?
Write to info@conclusionmp.com within 14 days of purchase and you get a full refund. No questions, no forms.
Ready when you are
The Hiring Desk, $12.
Instant download, one payment, every future revision included. Or take The Leadership Shelf, seven items for $45, and this kit works out at $6.43.