ficilcom/otame4-work-skills / skills/interview/interview-debrief
interview-debrief
VerifiedGo through an interview that has just finished and sort out what happened: the questions asked and how fully they were answered, which of them were prepared for and which came from the submitted documents, the reverse questions still unanswered, what the employer said about conditions and whether any of it was in writing, and when a result was promised. Use when a user has just come out of an interview and wants to record it while it is fresh or to work out what to fix before the next round; do not use to judge how it went, to estimate whether they will pass, or to read anything into the interviewer's manner.
| File | Kind | Size |
|---|---|---|
| SKILL.md | skill md | 5.4 KB |
| references/debrief-method.md | markdown | 5.6 KB |
| references/report-format.md | markdown | 2.1 KB |
| scripts/_common.py | script | 9.9 KB |
| scripts/review_interview.py | script | 10.8 KB |
scripts/review_interview.py
View on GitHub#!/usr/bin/env python3
"""Sort out what actually happened in an interview that has just finished.
聞かれた質問と答えられた度合い、逆質問の結果、企業側が口頭で述べた条件、次の
連絡の予定を整理する。準備していたのに答えられなかったものと、準備していなかった
ものを分ける。合否の見込みは推定しない。手応えを評価しない。
"""
from __future__ import annotations
from typing import Any
from _common import (
flag_collector,
optional_bool,
optional_choice,
optional_date,
optional_text,
require_list,
require_object,
require_text,
run_cli,
)
STAGES = ("casual", "first", "second", "final", "unknown")
ANSWER_LEVELS = ("full", "partial", "none")
REVERSE_RESULTS = ("answered", "partial", "deferred", "unanswered")
# 記憶が薄れる前に書き出す。これを超えて日が経った記録は、精度が落ちている前提で扱う。
FRESH_RECORD_DAYS = 3
def parse_questions(raw: object) -> list[dict[str, Any]]:
entries = require_list(raw if raw is not None else [], "questions")
questions = []
for index, entry in enumerate(entries):
item = require_object(entry, f"questions[{index}]")
answered = optional_choice(
item.get("answered"), f"questions[{index}].answered", ANSWER_LEVELS, "none"
)
questions.append(
{
"text": require_text(item.get("text"), f"questions[{index}].text"),
"answered": answered,
"prepared": optional_bool(item.get("prepared"), f"questions[{index}].prepared"),
"from_documents": bool(item.get("from_documents", False)),
"note": optional_text(item.get("note"), f"questions[{index}].note"),
}
)
return questions
def parse_reverse_questions(raw: object) -> list[dict[str, Any]]:
entries = require_list(raw if raw is not None else [], "questions_asked")
asked = []
for index, entry in enumerate(entries):
item = require_object(entry, f"questions_asked[{index}]")
result = optional_choice(
item.get("result"), f"questions_asked[{index}].result", REVERSE_RESULTS, "unanswered"
)
asked.append(
{
"text": require_text(item.get("text"), f"questions_asked[{index}].text"),
"result": result,
"note": optional_text(item.get("note"), f"questions_asked[{index}].note"),
}
)
return asked
def parse_statements(raw: object) -> list[dict[str, Any]]:
entries = require_list(raw if raw is not None else [], "employer_statements")
statements = []
for index, entry in enumerate(entries):
item = require_object(entry, f"employer_statements[{index}]")
statements.append(
{
"topic": require_text(item.get("topic"), f"employer_statements[{index}].topic"),
"statement": require_text(
item.get("statement"), f"employer_statements[{index}].statement"
),
"said_by": optional_text(item.get("said_by"), f"employer_statements[{index}].said_by"),
"in_writing": bool(item.get("in_writing", False)),
"conflicts_with_posting": optional_bool(
item.get("conflicts_with_posting"),
f"employer_statements[{index}].conflicts_with_posting",
),
"affects_conditions": bool(item.get("affects_conditions", False)),
}
)
return statements
def parse_next_steps(raw: object) -> dict[str, Any]:
steps = require_object(raw if raw is not None else {}, "next_steps")
next_stage = optional_choice(steps.get("next_stage"), "next_steps.next_stage", STAGES, "unknown")
return {
"next_stage": next_stage,
"result_promised_by": optional_date(
steps.get("result_promised_by"), "next_steps.result_promised_by"
),
"who_contacts": optional_text(steps.get("who_contacts"), "next_steps.who_contacts"),
}
def collect_flags(
questions: list[dict[str, Any]],
asked: list[dict[str, Any]],
statements: list[dict[str, Any]],
next_steps: dict[str, Any],
days_since: int | None,
) -> list[dict[str, Any]]:
flags, add = flag_collector()
if not questions:
add("questions_not_captured", "聞かれた質問が記録されていない。振り返りの材料がない")
prepared_but_missed = [
item["text"] for item in questions if item["prepared"] is True and item["answered"] != "full"
]
if prepared_but_missed:
add(
"prepared_but_not_answered",
"準備していたのに答えきれなかった質問がある。素材ではなく、話す順序や具体性の問題として見る",
prepared_but_missed,
)
unprepared = [
item["text"] for item in questions if item["prepared"] is False and item["answered"] != "full"
]
if unprepared:
add(
"not_prepared",
"準備していなかった質問で答えきれなかったものがある。次の面接までに素材を用意する",
unprepared,
)
from_documents = [
item["text"] for item in questions if item["from_documents"] and item["answered"] != "full"
]
if from_documents:
add(
"documents_not_defensible",
"提出書類の記述を聞かれて答えきれなかったものがある。書類の表現と事実のずれを先に直す",
from_documents,
)
unresolved = [item["text"] for item in asked if item["result"] in ("unanswered", "deferred", "partial")]
if unresolved:
add(
"reverse_questions_unresolved",
"逆質問のうち、答えが得られなかったものがある。次の段階か、条件提示のときに持ち越す",
unresolved,
)
if not asked:
add("no_reverse_questions_asked", "逆質問をしていない。確認したい論点が残っていないかを見る")
verbal_conditions = [
item["topic"]
for item in statements
if item["affects_conditions"] and not item["in_writing"]
]
if verbal_conditions:
add(
"conditions_stated_verbally",
"労働条件に関わる説明を口頭で受けている。書面で確認するまで確定した条件として扱わない",
verbal_conditions,
)
conflicts = [item["topic"] for item in statements if item["conflicts_with_posting"] is True]
if conflicts:
add(
"conflicts_with_posting",
"求人票の記載と食い違う説明を受けている。どちらが適用されるかを確認する",
conflicts,
)
if next_steps["result_promised_by"] is None:
add("result_timing_unknown", "結果の連絡時期を確認していない。いつまで待つかが決まらない")
if next_steps["who_contacts"] is None:
add("contact_unknown", "誰から連絡が来るかを確認していない")
if days_since is not None and days_since > FRESH_RECORD_DAYS:
add(
"recorded_late",
f"面接から{days_since}日経ってからの記録である。細部の再現度が落ちている前提で扱う",
)
return flags
def review(payload: object) -> dict[str, Any]:
data = require_object(payload, "input")
company = require_text(data.get("company"), "company")
stage = optional_choice(data.get("stage"), "stage", STAGES, "unknown")
interview_date = optional_date(data.get("date"), "date")
as_of = optional_date(data.get("as_of"), "as_of")
if interview_date and as_of and as_of < interview_date:
raise ValueError("as_of must not precede date")
days_since = (as_of - interview_date).days if interview_date and as_of else None
questions = parse_questions(data.get("questions"))
asked = parse_reverse_questions(data.get("questions_asked"))
statements = parse_statements(data.get("employer_statements"))
next_steps = parse_next_steps(data.get("next_steps"))
result_overdue = None
if next_steps["result_promised_by"] and as_of:
result_overdue = as_of > next_steps["result_promised_by"]
return {
"company": company,
"stage": stage,
"date": interview_date.isoformat() if interview_date else None,
"as_of": as_of.isoformat() if as_of else None,
"days_since_interview": days_since,
"summary": {
"questions": len(questions),
"answered_fully": sum(1 for item in questions if item["answered"] == "full"),
"answered_partially": sum(1 for item in questions if item["answered"] == "partial"),
"not_answered": sum(1 for item in questions if item["answered"] == "none"),
"from_documents": sum(1 for item in questions if item["from_documents"]),
"reverse_questions": len(asked),
"reverse_questions_resolved": sum(1 for item in asked if item["result"] == "answered"),
"employer_statements": len(statements),
"conditions_only_verbal": sum(
1 for item in statements if item["affects_conditions"] and not item["in_writing"]
),
},
"questions": questions,
"questions_asked": asked,
"employer_statements": statements,
"next_steps": {
"next_stage": next_steps["next_stage"],
"result_promised_by": (
next_steps["result_promised_by"].isoformat()
if next_steps["result_promised_by"]
else None
),
"who_contacts": next_steps["who_contacts"],
"result_overdue": result_overdue,
},
"carry_forward": {
"to_next_interview": [
item["text"] for item in questions if item["answered"] != "full"
]
+ [item["text"] for item in asked if item["result"] in ("unanswered", "deferred", "partial")],
"to_document_review": [
item["text"] for item in questions if item["from_documents"] and item["answered"] != "full"
],
"to_terms_check": [
item["topic"]
for item in statements
if item["affects_conditions"] and not item["in_writing"]
],
},
"flags": collect_flags(questions, asked, statements, next_steps, days_since),
"notes": [
"この出力は面接で起きたことの整理であり、手応えの評価でも合否の見込みでもない",
"口頭で受けた説明は、書面で確認するまで確定した労働条件ではない",
],
}
def main(argv: list[str] | None = None) -> int:
return run_cli(review, __doc__, argv)
if __name__ == "__main__":
raise SystemExit(main())
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| slack-gif-creatoranthropics/skillsKnowledge and utilities for creating animated GIFs optimized for Slack. Provides constraints, validation tools, and animation concepts. Use when users request animated GIFs for Slack like "make me a GIF of X doing Y for Slack." | Stars: 198k | Updated | Claude CodeCodex | Verified | Install |
| ponytailDietrichGebert/ponytailLazy senior dev mode for any coding task: the smallest change that fully solves it, and a reply a busy human understands in one read. | Stars: 194k | Updated | Claude CodeCodex | Verified | Install |
| ponytail-debtDietrichGebert/ponytailHarvest every shortcut: comment into one debt ledger, so deferrals get tracked instead of forgotten. One-shot report. | Stars: 194k | Updated | Claude CodeCodex | Verified | Install |
| ponytail-debtDietrichGebert/ponytailList every `shortcut:` comment (and older `ponytail:` ones) as a debt ledger. One-shot report, changes nothing. Use for "ponytail debt", "what did ponytail defer", "list the shortcuts", /ponytail-debt. | Stars: 194k | Updated | Claude CodeCodex | Verified | Install |