prepare-candidate-interviews

name	prepare-candidate-interviews
description	Build a structured candidate interview plan, interviewer assignments, candidate brief, questions, probes, and scorecards from an accepted role definition. Use before interviews or to repair an inconsistent interview loop. Do not make hiring decisions or expose inappropriate sensitive information.

Prepare Candidate Interviews

Design an interview loop that gathers decision-relevant evidence consistently and gives the candidate a fair, coherent experience. Start with the material needed for the next conversation and expand to the whole loop only when that is the user's job.

1. Resolve the role, candidate, and interview

Establish whether the user is designing a complete interview loop, preparing interviewers for a scheduled stage, or repairing an inconsistent process. Resolve the candidate, role, decision stage, event or events, timing, interviewers, and intended reviewers before combining records.

When the request refers to an upcoming, next, or named interview, start from the calendar if it is available. Match the event using candidate and interviewer names, email addresses, role, stage, company domain, and process context. Ask when the match is unclear rather than attaching the wrong candidate or private material.

If Calendar is not connected, explain briefly that it improves event, participant, and timing resolution, then continue from an open invite, ATS schedule, meeting link, or details the user provides. Do not make the connection a prerequisite.

Calibrate depth to the time remaining and stakes. If an interview is imminent, deliver the essential interviewer brief, evidence gaps, and questions first. With more time, build or repair the full loop and supporting materials.

2. Gather the accepted hiring context

Use the accepted job description or internal role profile, success outcomes, scorecard, candidate materials, application and sourcing evidence, prior-stage notes, calendar context, interviewer availability, process policy, and accommodation information the user is authorized to use.

Keep the accepted hiring criteria distinct from candidate-specific evidence. Identify competencies already supported, contradicted, or still untested so the next stage does not repeat questions without purpose. Flag missing, stale, or inconsistent role criteria and prior feedback rather than quietly designing around them.

Use only job-related, authorized information. Never introduce protected characteristics, medical details, family status, unrelated personal information, or unsupported inferences into the evaluation.

3. Design evidence coverage

Define what the stage must learn and assign clear ownership. Build the plan around:

• the interview purpose and candidate-facing agenda;
• competency and evidence ownership by interviewer;
• structured core questions with job-related probes;
• candidate-specific areas to verify, separated from the standard question set;
• scoring anchors and evidence-capture fields;
• work samples or exercises only when proportionate and approved;
• handoffs, timing, and debrief expectations; and
• accessibility, privacy, consistency, and candidate-experience checks.

Use the fewest stages and questions that can produce the evidence the decision needs. Do not create extra interviews to compensate for unclear ownership, and do not tailor away the common evidence bar merely because one candidate has a different background.

4. Prepare the panel and candidate materials

Give each interviewer a concise brief with the stage purpose, context they need, evidence they own, questions and probes, scoring anchors, topics already covered, and what to record. Keep private notes, unsupported concerns, and information irrelevant to that interviewer out of the brief.

Prepare a separate candidate-facing agenda or instructions when useful. Make the format, timing, participants, preparation, accessibility path, and expectations clear without exposing internal evaluation notes or other candidates' information.

Keep source links or references close to candidate-specific claims the interviewer may need to verify. Distinguish facts, candidate claims, prior interviewer judgment, and companion inference.

5. Check the loop before it runs

Verify that every must-have has an owner, duplicated questioning is intentional, scoring anchors are usable, the schedule is realistic, and the loop can distinguish evidence from impression. Check participant identity, time zones, meeting links, handoffs, accommodations the team is authorized to act on, and the planned debrief.

Deliver the interview plan, interviewer assignments, panel briefs, candidate-facing material, and a short list of unresolved decisions or evidence gaps. Let the hiring team correct the criteria, coverage, timing, questions, tone, and sharing boundaries before anything is scheduled or sent.

6. Run the handoff within scope

Follow Strawberry's active scoped permission for every calendar, ATS account, candidate, document, destination, and action. Creating a draft, sharing panel materials, sending candidate information, scheduling or changing interviews, changing ATS records, and communicating externally are separate actions. Draft or ask when permission is insufficient, stop when identity, scope, impact, or sensitive-data handling changes, and verify completed external actions.

After authorized interviews, use Strawberry's meeting capture or another reliable record, then use strawberry/recruiting/synthesize-interview-feedback to map evidence, contradictions, and unknowns to the accepted scorecard. A qualified human owns advancement, rejection, offer, and hiring decisions.

After the team accepts the interview kit, offer to save its criteria, coverage map, questions, scorecards, brief shape, and review points as a custom or team skill. Add a Routine only for stable preparation or follow-up with a real trigger, approved sources, draft-first output, destination, review behavior, and stop conditions—never for the hiring decision.