risk-register — скилл для Claude Code: реестр рисков проекта
Скилл risk-register создаёт качественный реестр рисков проекта с матрицей вероятность × последствие, назначает владельцев, описывает стратегии смягчения и планы на случай срабатывания.
MIT ★ 1 627 обновлён 2026-10-25
npx skills add microsoft/hve-core --skill risk-register -a claude-codeЧто делает risk-register
Скилл собирает контекст проекта — сроки, стейкхолдеры, зависимости, известные проблемы — и генерирует два документа в docs/risks/: risk-register.md со сводной таблицей и детальными карточками рисков и risk-mitigation-plan.md с приоритетными действиями. Каждый риск описывается по формуле «причина → событие → последствие», получает оценку вероятности и воздействия (Low, Medium, High), числовой балл для сортировки, триггеры и назначенного ответственного.
Скилл требует ручной проверки: он помечен CAUTION и прямо указывает, что сгенерированные оценки должны пройти через квалифицированных специалистов по управлению рисками. Модель не вызывается автоматически (disable-model-invocation: true). Можно передать project-name и focus-area, чтобы сузить область анализа.
Какие задачи решает
- структурировать выявление и документирование рисков проекта
- расставить приоритеты по матрице вероятность × последствие
- назначить владельцев и триггеры для каждого риска
- подготовить план реагирования на высокоприоритетные угрозы
- создать версионированные снимки реестра по датам
Примеры запросов агенту
- Создай реестр рисков для проекта
api-gateway - Оцени риски в области безопасности
- Обнови
docs/risks/risk-register.mdс учётом новых зависимостей
Как установить risk-register
1Через Skills CLI (Vercel)
npx skills add microsoft/hve-core --skill risk-register -a claude-code2Вручную: положить файлы
~/.claude/skills/risk-register/SKILL.mdПапка скилла целиком; для одного проекта — .claude/skills/risk-register/
1Через Skills CLI (Vercel)
npx skills add microsoft/hve-core --skill risk-register -a codex2Вручную: положить файлы
.agents/skills/risk-register/SKILL.mdДля всех проектов — ~/.agents/skills/
1Через Skills CLI (Vercel)
npx skills add microsoft/hve-core --skill risk-register -a cursor2Вручную: положить файлы
.cursor/skills/risk-register/SKILL.mdCursor читает и .agents/skills/, и .claude/skills/
1GitHub CLI
gh skill install microsoft/hve-core .github/skills/security/risk-register2Вручную: положить файлы
.github/skills/risk-register/SKILL.md1Через Skills CLI (Vercel)
npx skills add microsoft/hve-core --skill risk-register -a gemini-cli2Командой платформы
gemini skills install https://github.com/microsoft/hve-core --path .github/skills/security/risk-registerВручную: положить файлы
.devin/skills/risk-register/SKILL.mdСтарый путь .windsurf/skills/ тоже читается
Claude.ai
Упакуйте папку в zip и загрузите в Settings → Capabilities → Skills
По шагам, с командами для каждой платформы: Как установить скилл в Claude Code: пять способов
Вопросы о risk-register
Как установить risk-register в Claude Code?
Выполни npx skills add microsoft/hve-core --skill risk-register -a claude-code. Скилл окажется в ~/.claude/skills/risk-register/SKILL.md.
Что делает risk-register?
Генерирует реестр рисков проекта с качественной оценкой по матрице вероятность × последствие, создаёт план смягчения и назначает ответственных. Документы попадают в docs/risks/.
Можно ли использовать сгенерированный реестр без проверки?
Нет. Скилл помечен CAUTION и явно требует проверки квалифицированными специалистами: ИИ может пропустить критичные риски или дать некорректные оценки.
Исходный текст
SKILL.md · MIT
--- name: risk-register description: 'Create a qualitative risk register and mitigation plan using a Probability x Impact matrix. Use when a project needs structured risk identification, scoring, ownership, and response planning.' argument-hint: '[project-name=...] [focus-area=...]' license: MIT user-invocable: true disable-model-invocation: true --- # Risk Register Generator > [!CAUTION] > This skill is an **assistive tool only** and does not replace professional security and risk management tooling or qualified human review. All generated risk registers, risk scores, and mitigation strategies must be reviewed and validated by qualified security and risk professionals before use. AI outputs may contain inaccuracies, miss critical risks, or produce assessments that are incomplete or inappropriate for your environment. ## Purpose and Role Act as a risk management assistant. Help the user identify, document, and prioritize project risks using a qualitative risk assessment approach based on a Probability × Impact (P×I) risk matrix. Use clear, simple, professional language and avoid unnecessary detail. Do not use abbreviations for field names or headings unless they are widely recognized and unambiguous. Place all outputs in the `docs/risks/` folder. ## Inputs * `project-name` (optional): Project name used for context and, in multi-project repositories, file naming. * `focus-area` (optional): Risk focus area, domain, component, milestone, or concern to emphasize. ## Step 1: Gather Project Context If not already available in the repository, ask the user to provide: * Project name and short description. * Timeline and key milestones. * Stakeholders and dependencies. * Technical components or systems involved. * Known risks or concerns. * Sources of uncertainty. * Assessment of potential consequences related to project objectives. ## Step 2: Prepare Risk Documentation Structure * Ensure the folder `docs/risks/` exists. Create it if missing. * Place all generated files inside the `docs/risks/` folder. * Use clear and direct file names and headings. File naming conventions: * Primary register: `risk-register.md`. * Versioned register snapshot: `risk-register-YYYY-MM-DD.md`. * Mitigation plan: `risk-mitigation-plan.md`. * Multi-project repository register: `risk-register-[project-name].md`. ## Step 3: Create `risk-register.md` in `docs/risks/` Include these sections: * Executive Summary. * Project Overview. * Risk Assessment Methodology. * Overview Table of Risks. * Detailed Risk Entries. ### Risk Assessment Methodology Document this methodology in the register: * Risks are scored using a P×I matrix with qualitative bands (Low, Medium, High) for both Probability and Impact. * Default Probability scale: * Low: unlikely. * Medium: possible. * High: likely. * Default Impact scale: * Low: minor effect. * Medium: moderate effect. * High: major effect. * Risk Score (Qualitative) = Probability × Impact using qualitative labels, such as High × Medium. * Risk Score (Numeric) converts qualitative ratings to numbers for sorting: * Low = 1. * Medium = 2. * High = 3. * Example: High × Medium = 3 × 2 = 6. * Document rationale for each rating in 1 to 2 lines for consistency. ### Overview Table of Risks Use these columns: * Risk ID. * Risk Title. * Description (Cause → Event → Impact). * Probability (Low, Medium, High). * Impact (Low, Medium, High). * Risk Score (Qualitative) = Probability × Impact, such as High × Medium. * Risk Score (Numeric) = Numeric value for sorting, such as 6. Example: | Risk ID | Risk Title | Description (Cause → Event → Impact) | Probability | Impact | Risk Score (Qualitative) | Risk Score (Numeric) | |---------|----------------------------|-----------------------------------------------------------------------------------------------------------------------------------------|-------------|--------|--------------------------|----------------------| | R-001 | API rate limits exceeded | High request volume without effective throttling → API rate limits are exceeded → Requests fail and downstream workflows degrade | High | Medium | High × Medium | 6 | | R-002 | Key developer unavailable | Single point of knowledge in a key area → Key developer becomes unavailable → Delivery slows and defects increase | Medium | High | Medium × High | 6 | | R-003 | Third-party service outage | Dependency on an external provider → Third-party service becomes unavailable → Features relying on it fail and user experience degrades | Medium | Medium | Medium × Medium | 4 | ### Detailed Risk Entries Include these fields for each risk: * Risk ID and Title. * Description (Cause → Event → Impact). * Probability and Impact ratings plus rationale. * Risk Score (Qualitative) = Probability × Impact, such as High × Medium. * Risk Score (Numeric), such as 6. * Category. * Mitigation Strategy. * Contingency Plan. * Trigger Events. * Owner. * Status. Use short, focused descriptions. Avoid jargon and unnecessary elaboration. Sort all risks by descending Risk Score (Numeric) to highlight the most critical risks. ## Step 4: Create `risk-mitigation-plan.md` in `docs/risks/` Base the mitigation plan on the mitigation strategies already defined in `risk-register.md`. Focus on the highest priority risks, especially those with high probability and high impact, and summarize the planned responses. Use this outline: * Top Priority Risks (High Uncertainty and High Consequence). * Risk Response Actions derived from mitigation strategies in `risk-register.md`. * Resource Requirements. * Communication Plan. * Risk Reassessment Schedule. ## Guidelines * Use Cause → Event → Impact format for risk statements. * Define and document qualitative scales upfront. * Record rationale for each rating. * Include trigger events and assign a single accountable owner per risk. * Establish reassessment cadence and closure criteria. * Use clear, concise, and simple language throughout all sections. * Avoid unnecessary detail or verbosity. * Avoid abbreviations for field names or headings, for example use "Priority" instead of "P", unless they are widely recognized and unambiguous. * Include both technical and non-technical risks. * Focus on actionable mitigation strategies. * Consider internal and external risk factors. ## Stop Rules * Ask for missing project context when repository evidence and user input do not provide enough information to identify meaningful risks. * Stop before writing tracker-bound or externally visible risk artifacts outside `docs/risks/`. * Stop before treating generated risks, scores, or mitigations as final without qualified human review. ## Success Criteria * `docs/risks/` contains the generated risk register and mitigation plan. * The risk register defines the qualitative Probability × Impact scoring method and rating rationale. * The overview table and detailed entries use the required fields. * Risks are sorted by descending numeric score. * The mitigation plan is derived from the register and focuses on the highest priority risks.
© microsoft · MIT · microsoft/hve-core