{"id":"ccfba71e-940e-4ed2-979d-eef840240052","shortId":"AcqjKQ","kind":"skill","title":"conducting-interviews","tagline":"Run structured behavioral HIRING interviews: plan, questions, scorecard, debrief. See also: conducting-user-interviews (product research).","description":"# Conducting Interviews (Structured, Behavioral)\n\n## Scope\n\n**Covers**\n- Preparing and running structured interviews (screen + loop) with consistent criteria\n- Behavioral interviewing mapped to competencies/values\n- Getting to “substance over polish” (avoiding “confident but shallow” signal)\n- Capturing evidence, scoring consistently, and writing a debrief-ready summary\n\n**When to use**\n- “Help me conduct interviews for a <role>.”\n- “Create an interview script / interview question set / scorecard for <role>.”\n- “Design an interview loop and structured rubric for <role>.”\n- “Improve interviewer consistency and reduce bias.”\n\n**When NOT to use**\n- You need to define the role outcomes or write the job description (use `writing-job-descriptions` first)\n- You need to make a final hiring decision, design work samples, or run reference checks (use `evaluating-candidates`)\n- You need to conduct user/customer research interviews (use `conducting-user-interviews` — completely different skill)\n- You need to onboard the person after hiring (use `onboarding-new-hires`)\n- You need legal/HR compliance guidance or to adjudicate complex employment risk (this skill is not legal advice)\n- You need compensation/offer strategy or negotiation coaching\n\n## Inputs\n\n**Minimum required**\n- Role + level + function (e.g., “Senior PM”, “Engineering Manager”)\n- Interview stage(s) to design/run (screen, hiring manager, panel, etc.) + duration(s)\n- Evaluation criteria: 4–8 competencies/values to measure (or your existing rubric)\n- Company/team context candidates should know (mission, what’s hard, why now)\n- Candidate materials (resume/portfolio) + any areas to probe\n\n**Missing-info strategy**\n- Ask up to 5 questions from [references/INTAKE.md](references/INTAKE.md).\n- If criteria aren’t provided, propose a default criteria set and clearly label it as an assumption.\n\n## Outputs (deliverables)\n\nProduce an **Interview Execution Pack** in Markdown (in-chat; or as files if requested):\n\n1) **Interview plan** (stage purpose, criteria, agenda, timeboxes)\n2) **Question map** (questions → competency/value → what good looks like → follow-up probes)\n3) **Interviewer script** (opening, transitions, probes, close)\n4) **Notes + scorecard** (rating anchors + evidence capture)\n5) **Debrief summary template** (evidence-based strengths/concerns + hire/no-hire signal + follow-ups)\n6) **Risks / Open questions / Next steps** (always included)\n\nTemplates: [references/TEMPLATES.md](references/TEMPLATES.md)  \nExpanded guidance: [references/WORKFLOW.md](references/WORKFLOW.md)\n\n## Workflow (7 steps)\n\n### 1) Intake + define the stage\n- **Inputs:** user request; [references/INTAKE.md](references/INTAKE.md).\n- **Actions:** Confirm role, stage(s), duration, and who else interviews. Identify must-measure criteria and any “must not” red flags.\n- **Outputs:** Interview brief + assumptions/unknowns list.\n- **Checks:** You can state the stage goal in one sentence (e.g., “screen for X; sell Y; decide Z”).\n\n### 2) Lock evaluation criteria (don’t improvise later)\n- **Inputs:** competencies/values; role context.\n- **Actions:** Choose 4–8 criteria; define 1–2 “strong” and “weak” anchors per criterion. Ensure each criterion is observable via evidence.\n- **Outputs:** Criteria table with anchors.\n- **Checks:** Every criterion has a definition + evidence hints; no criterion is “vibe”.\n\n### 3) Build the question map (behavioral first)\n- **Inputs:** criteria table.\n- **Actions:** Write 1–2 primary questions per criterion (behavioral: “tell me about a time…”). Add probes that force specifics (role, constraints, trade-offs, results, what you’d do differently). Add two global questions: “How did you prepare?” and “Why here?”\n- **Outputs:** Question map table.\n- **Checks:** Each question maps to exactly one primary criterion; no double-barreled questions.\n\n### 4) Write the interviewer script (runbook)\n- **Inputs:** question map; timeboxes.\n- **Actions:** Assemble an interview flow: opening (set context + structure), question sequence, note-taking reminders, and a consistent close: “Is there anything else you want to make sure we covered?”\n- **Outputs:** Interviewer script with timestamps.\n- **Checks:** Script fits in time; includes “sell” moments appropriate to stage; includes candidate questions time.\n\n### 5) Prepare for “substance over polish”\n- **Inputs:** question map; candidate materials.\n- **Actions:** Add “substance checks” for polished communicators (ask for concrete examples, counterfactuals, and specific decisions). Add “structure help” for less polished candidates (rephrase, clarify what’s being asked) without leading.\n- **Outputs:** Substance-vs-delivery guardrails embedded in the script.\n- **Checks:** The plan reduces false positives from confident delivery and false negatives from imperfect structure.\n\n### 6) Score using evidence (immediately after)\n- **Inputs:** notes; scorecard template.\n- **Actions:** Fill the scorecard with evidence snippets before discussing with others. Rate each criterion with anchors. Write a 5–8 sentence evidence-based summary and list follow-up questions.\n- **Outputs:** Completed notes + scorecard + summary.\n- **Checks:** Every rating has supporting evidence; the overall recommendation is consistent with criterion ratings.\n\n### 7) Debrief + quality gate + finalize pack\n- **Inputs:** completed scorecard; debrief template.\n- **Actions:** Produce the debrief-ready packet; run [references/CHECKLISTS.md](references/CHECKLISTS.md) and score with [references/RUBRIC.md](references/RUBRIC.md). Include Risks/Open questions/Next steps.\n- **Outputs:** Final Interview Execution Pack.\n- **Checks:** Clear recommendation + uncertainty; fair process; next steps defined (additional interview, reference check, work sample, etc.).\n\n## Quality gate (required)\n- Use [references/CHECKLISTS.md](references/CHECKLISTS.md) and [references/RUBRIC.md](references/RUBRIC.md).\n- Always include: **Risks**, **Open questions**, **Next steps**.\n\n## Examples\n\n**Example 1 (Screen):** “Create a 30-minute phone screen for a Senior Product Manager. I want to evaluate product sense, execution, and collaboration. Output the Interview Execution Pack with a question map and scorecard.”  \nExpected: timeboxed script, behavioral questions, clear anchors, and a scorecard that captures evidence.\n\n**Example 2 (Loop):** “Design a structured interview loop for a Staff Engineer, including a hiring manager interview and a cross-functional panel. Map questions to our values and include a debrief template.”  \nExpected: stage goals, consistent criteria across interviewers, and artifacts that make debriefs evidence-based.\n\n**Boundary example (redirect):** “Help me decide which of these 3 candidates to hire based on their interview notes and references.”\nResponse: redirect to `evaluating-candidates` — this skill designs and runs interviews, it does not synthesize cross-candidate hiring decisions.\n\n**Boundary example (wrong domain):** “I need to interview 10 users about their onboarding experience with our product.”\nResponse: redirect to `conducting-user-interviews` — this skill is for hiring interviews, not user/customer research.\n\n**Boundary example (missing criteria):** “Just tell me if this candidate is good; I don’t have criteria or notes.”\nResponse: require criteria + evidence; propose default criteria and ask the user to paste notes or run a structured interview first.\n\n## Anti-patterns (common failure modes)\n\n1. **”Wing it” interviewing** — Skipping structured criteria and question maps, then relying on gut feel. This produces inconsistent signals and legal risk. Always lock criteria before writing questions.\n2. **Brainteaser / hypothetical-only questions** — Asking “How many golf balls fit in a school bus?” or purely hypothetical scenarios instead of behavioral evidence. These predict interview prep, not job performance.\n3. **Halo/horns from first 5 minutes** — Letting initial rapport (or lack thereof) color all subsequent scoring. Mitigate by scoring each criterion independently with evidence before writing an overall summary.\n4. **Identical questions for every role** — Reusing the same generic question bank regardless of role, level, or competency. Every question should map to a specific criterion for this specific role.\n5. **Skipping the “substance over polish” check** — Rewarding confident, articulate delivery while penalizing candidates who need a moment to organize their thoughts. Always include specificity probes and structural support for less polished communicators.","tags":["conducting","interviews","lenny","skills","plus","liqiongyu","agent-skills","ai-agents","automation","claude","codex","prompt-engineering"],"capabilities":["skill","source-liqiongyu","skill-conducting-interviews","topic-agent-skills","topic-ai-agents","topic-automation","topic-claude","topic-codex","topic-prompt-engineering","topic-refoundai","topic-skillpack"],"categories":["lenny_skills_plus"],"synonyms":[],"warnings":[],"endpointUrl":"https://skills.sh/liqiongyu/lenny_skills_plus/conducting-interviews","protocol":"skill","transport":"skills-sh","auth":{"type":"none","details":{"cli":"npx skills add liqiongyu/lenny_skills_plus","source_repo":"https://github.com/liqiongyu/lenny_skills_plus","install_from":"skills.sh"}},"qualityScore":"0.474","qualityRationale":"deterministic score 0.47 from registry signals: · indexed on github topic:agent-skills · 49 github stars · SKILL.md body (8,546 chars)","verified":false,"liveness":"unknown","lastLivenessCheck":null,"agentReviews":{"count":0,"score_avg":null,"cost_usd_avg":null,"success_rate":null,"latency_p50_ms":null,"narrative_summary":null,"summary_updated_at":null},"enrichmentModel":"deterministic:skill-github:v1","enrichmentVersion":1,"enrichedAt":"2026-04-22T06:56:19.542Z","embedding":null,"createdAt":"2026-04-18T22:16:22.158Z","updatedAt":"2026-04-22T06:56:19.542Z","lastSeenAt":"2026-04-22T06:56:19.542Z","tsv":"'1':286,352,424,468,780,993 '10':923 '2':294,406,425,469,827,1021 '3':307,456,883,1052 '30':784 '4':213,314,420,525,1081 '5':247,321,585,679,1056,1111 '6':334,651 '7':350,711 '8':214,421,680 'across':864 'action':362,418,466,535,596,661,722 'add':480,496,597,611 'addit':755 'adjud':171 'advic':180 'agenda':292 'also':14 'alway':340,771,1015,1133 'anchor':318,429,443,676,819 'anti':988 'anti-pattern':987 'anyth':556 'appropri':578 'area':237 'aren':254 'articul':1120 'artifact':867 'ask':244,603,623,975,1027 'assembl':536 'assumpt':268 'assumptions/unknowns':386 'avoid':47 'ball':1031 'bank':1092 'barrel':523 'base':327,684,873,887 'behavior':6,24,37,461,474,816,1043 'bias':94 'boundari':874,915,948 'brainteas':1022 'brief':385 'build':457 'bus':1036 'candid':135,224,233,582,594,617,884,899,912,957,1124 'captur':52,320,824 'chat':280 'check':131,388,444,511,570,599,636,697,746,758,1117 'choos':419 'clarifi':619 'clear':263,747,818 'close':313,553 'coach':187 'collabor':801 'color':1064 'common':990 'communic':602,1143 'company/team':222 'compensation/offer':183 'compet':1098 'competencies/values':41,215,415 'competency/value':298 'complet':148,693,718 'complex':172 'complianc':167 'concret':605 'conduct':2,16,21,68,139,145,936 'conducting-interview':1 'conducting-user-interview':15,144,935 'confid':48,643,1119 'confirm':363 'consist':35,55,91,552,707,862 'constraint':486 'context':223,417,542 'counterfactu':607 'cover':26,564 'creat':72,782 'criteria':36,212,253,260,291,376,409,422,440,464,863,951,964,969,973,999,1017 'criterion':431,434,446,453,473,519,674,709,1072,1106 'cross':846,911 'cross-candid':910 'cross-funct':845 'd':493 'debrief':12,60,322,712,720,726,857,870 'debrief-readi':59,725 'decid':404,879 'decis':124,610,914 'default':259,972 'defin':102,354,423,754 'definit':449 'deliver':270 'deliveri':630,644,1121 'descript':110,115 'design':81,125,829,902 'design/run':203 'differ':149,495 'discuss':669 'domain':918 'doubl':522 'double-barrel':521 'durat':209,367 'e.g':194,398 'els':370,557 'embed':632 'employ':173 'engin':197,837 'ensur':432 'etc':208,761 'evalu':134,211,408,796,898 'evaluating-candid':133,897 'everi':445,698,1085,1099 'evid':53,319,326,438,450,654,666,683,702,825,872,970,1044,1075 'evidence-bas':325,682,871 'exact':516 'exampl':606,778,779,826,875,916,949 'execut':274,744,799,805 'exist':220 'expand':345 'expect':813,859 'experi':928 'failur':991 'fair':750 'fals':640,646 'feel':1007 'file':283 'fill':662 'final':122,715,742 'first':116,462,986,1055 'fit':572,1032 'flag':382 'flow':539 'follow':304,332,689 'follow-up':303,331,688 'forc':483 'function':193,847 'gate':714,763 'generic':1090 'get':42 'global':498 'goal':394,861 'golf':1030 'good':300,959 'guardrail':631 'guidanc':168,346 'gut':1006 'halo/horns':1053 'hard':230 'help':66,613,877 'hint':451 'hire':7,123,158,163,205,840,886,913,943 'hire/no-hire':329 'hypothet':1024,1039 'hypothetical-on':1023 'ident':1082 'identifi':372 'immedi':655 'imperfect':649 'improv':89 'improvis':412 'in-chat':278 'includ':341,575,581,737,772,838,855,1134 'inconsist':1010 'independ':1073 'info':242 'initi':1059 'input':188,357,414,463,531,591,657,717 'instead':1041 'intak':353 'interview':3,8,18,22,31,38,69,74,76,83,90,142,147,199,273,287,308,371,384,528,538,566,743,756,804,832,842,865,890,905,922,938,944,985,996,1047 'job':109,114,1050 'know':226 'label':264 'lack':1062 'later':413 'lead':625 'legal':179,1013 'legal/hr':166 'less':615,1141 'let':1058 'level':192,1096 'like':302 'list':387,687 'lock':407,1016 'look':301 'loop':33,84,828,833 'make':120,561,869 'manag':198,206,792,841 'mani':1029 'map':39,296,460,509,514,533,593,810,849,1002,1102 'markdown':277 'materi':234,595 'measur':217,375 'minimum':189 'minut':785,1057 'miss':241,950 'missing-info':240 'mission':227 'mitig':1068 'mode':992 'moment':577,1128 'must':374,379 'must-measur':373 'need':100,118,137,152,165,182,920,1126 'negat':647 'negoti':186 'new':162 'next':338,752,776 'note':315,547,658,694,891,966,980 'note-tak':546 'observ':436 'off':489 'onboard':154,161,927 'onboarding-new-hir':160 'one':396,517 'open':310,336,540,774 'organ':1130 'other':671 'outcom':105 'output':269,383,439,507,565,626,692,741,802 'overal':704,1079 'pack':275,716,745,806 'packet':728 'panel':207,848 'past':979 'pattern':989 'penal':1123 'per':430,472 'perform':1051 'person':156 'phone':786 'plan':9,288,638 'pm':196 'polish':46,590,601,616,1116,1142 'posit':641 'predict':1046 'prep':1048 'prepar':27,503,586 'primari':470,518 'probe':239,306,312,481,1136 'process':751 'produc':271,723,1009 'product':19,791,797,931 'propos':257,971 'provid':256 'pure':1038 'purpos':290 'qualiti':713,762 'question':10,77,248,295,297,337,459,471,499,508,513,524,532,544,583,592,691,775,809,817,850,1001,1020,1026,1083,1091,1100 'questions/next':739 'rapport':1060 'rate':317,672,699,710 'readi':61,727 'recommend':705,748 'red':381 'redirect':876,895,933 'reduc':93,639 'refer':130,757,893 'references/checklists.md':730,731,766,767 'references/intake.md':250,251,360,361 'references/rubric.md':735,736,769,770 'references/templates.md':343,344 'references/workflow.md':347,348 'regardless':1093 'reli':1004 'remind':549 'rephras':618 'request':285,359 'requir':190,764,968 'research':20,141,947 'respons':894,932,967 'result':490 'resume/portfolio':235 'reus':1087 'reward':1118 'risk':174,335,773,1014 'risks/open':738 'role':104,191,364,416,485,1086,1095,1110 'rubric':87,221 'run':4,29,129,729,904,982 'runbook':530 'sampl':127,760 'scenario':1040 'school':1035 'scope':25 'score':54,652,733,1067,1070 'scorecard':11,79,316,659,664,695,719,812,822 'screen':32,204,399,781,787 'script':75,309,529,567,571,635,815 'see':13 'sell':402,576 'senior':195,790 'sens':798 'sentenc':397,681 'sequenc':545 'set':78,261,541 'shallow':50 'signal':51,330,1011 'skill':150,176,901,940 'skill-conducting-interviews' 'skip':997,1112 'snippet':667 'source-liqiongyu' 'specif':484,609,1105,1109,1135 'staff':836 'stage':200,289,356,365,393,580,860 'state':391 'step':339,351,740,753,777 'strategi':184,243 'strengths/concerns':328 'strong':426 'structur':5,23,30,86,543,612,650,831,984,998,1138 'subsequ':1066 'substanc':44,588,598,628,1114 'substance-vs-deliveri':627 'summari':62,323,685,696,1080 'support':701,1139 'sure':562 'synthes':909 'tabl':441,465,510 'take':548 'tell':475,953 'templat':324,342,660,721,858 'thereof':1063 'thought':1132 'time':479,574,584 'timebox':293,534,814 'timestamp':569 'topic-agent-skills' 'topic-ai-agents' 'topic-automation' 'topic-claude' 'topic-codex' 'topic-prompt-engineering' 'topic-refoundai' 'topic-skillpack' 'trade':488 'trade-off':487 'transit':311 'two':497 'uncertainti':749 'up':333 'use':65,98,111,132,143,159,653,765 'user':17,146,358,924,937,977 'user/customer':140,946 'valu':853 'via':437 'vibe':455 'vs':629 'want':559,794 'weak':428 'wing':994 'without':624 'work':126,759 'workflow':349 'write':57,107,113,467,526,677,1019,1077 'writing-job-descript':112 'wrong':917 'x':401 'y':403 'z':405","prices":[{"id":"2198a184-c57d-4936-a67c-64abe83c1b31","listingId":"ccfba71e-940e-4ed2-979d-eef840240052","amountUsd":"0","unit":"free","nativeCurrency":null,"nativeAmount":null,"chain":null,"payTo":null,"paymentMethod":"skill-free","isPrimary":true,"details":{"org":"liqiongyu","category":"lenny_skills_plus","install_from":"skills.sh"},"createdAt":"2026-04-18T22:16:22.158Z"}],"sources":[{"listingId":"ccfba71e-940e-4ed2-979d-eef840240052","source":"github","sourceId":"liqiongyu/lenny_skills_plus/conducting-interviews","sourceUrl":"https://github.com/liqiongyu/lenny_skills_plus/tree/main/skills/conducting-interviews","isPrimary":false,"firstSeenAt":"2026-04-18T22:16:22.158Z","lastSeenAt":"2026-04-22T06:56:19.542Z"}],"details":{"listingId":"ccfba71e-940e-4ed2-979d-eef840240052","quickStartSnippet":null,"exampleRequest":null,"exampleResponse":null,"schema":null,"openapiUrl":null,"agentsTxtUrl":null,"citations":[],"useCases":[],"bestFor":[],"notFor":[],"kindDetails":{"org":"liqiongyu","slug":"conducting-interviews","github":{"repo":"liqiongyu/lenny_skills_plus","stars":49,"topics":["agent-skills","ai-agents","automation","claude","codex","prompt-engineering","refoundai","skillpack"],"license":"apache-2.0","html_url":"https://github.com/liqiongyu/lenny_skills_plus","pushed_at":"2026-04-04T06:30:11Z","description":"86 agent-executable skill packs converted from RefoundAI’s Lenny skills (unofficial). Works with Codex + Claude Code.","skill_md_sha":"a25fea203446662913a47c2faf087ac13b236c64","skill_md_path":"skills/conducting-interviews/SKILL.md","default_branch":"main","skill_tree_url":"https://github.com/liqiongyu/lenny_skills_plus/tree/main/skills/conducting-interviews"},"layout":"multi","source":"github","category":"lenny_skills_plus","frontmatter":{"name":"conducting-interviews","description":"Run structured behavioral HIRING interviews: plan, questions, scorecard, debrief. See also: conducting-user-interviews (product research)."},"skills_sh_url":"https://skills.sh/liqiongyu/lenny_skills_plus/conducting-interviews"},"updatedAt":"2026-04-22T06:56:19.542Z"}}