product-legal — плагин для Claude Code: что делает и как установить
Плагин product-legal для Claude Code проводит юридический анализ релизов, рекламных текстов и продуктовых рисков по внутренней истории компании.
Apache-2.0 ★ 10 060 обновлён 2026-10-23
/plugin marketplace add anthropics/claude-for-legal
/plugin install product-legal@claude-for-legalЧто делает product-legal
Плагин автоматизирует первичную юридическую проверку релизов, разбирает маркетинговые формулировки и оценивает риски функций. Работа опирается на таблицу калибровки рисков, которую инструмент собирает при первом запуске на основе прошлых обзоров компании. Конфигурация сохраняется в ~/.claude/plugins/config/claude-for-legal/product-legal/CLAUDE.md и сохраняется после обновления плагина. Для работы подключаются интеграции вроде Slack, Google Drive, Linear, Atlassian и Asana.
Каждый результат работы создается как черновик для проверки юристом. Без подключенного инструмента поиска цитирования помечаются тегом [verify]. Конфигурацию рисков нужно перезапускать при изменении политики компании.
Какие задачи решает
- проверить релиз продукта по фреймворку компании
- оценить риски конкретной функции
- проверить маркетинговые утверждения перед запуском
- быстро ответить на юридический вопрос по продукту
Примеры запросов агенту
- /product-legal:cold-start-interview
- /product-legal:is-this-a-problem "Can we A/B test the pricing page?"
- /product-legal:launch-review PROJ-1234
Как установить product-legal
Через маркетплейс плагинов
/plugin marketplace add anthropics/claude-for-legal/plugin install product-legal@claude-for-legalПо шагам, с командами для каждой платформы: Плагины Claude Code: маркетплейсы, установка и как собрать свой
Вопросы о product-legal
Как установить product-legal в Claude Code?
Выполните команды /plugin marketplace add anthropics/claude-for-legal и /plugin install product-legal@claude-for-legal.
Что делает product-legal?
Плагин проверяет продуктовые запуски и маркетинговые тексты на соответствие калибровке рисков компании, готовит черновики юридических заключений и автоматизирует первичную триажу запросов.
Где хранится конфигурация плагина?
Конфигурация и таблица калибровки рисков сохраняются в файле ~/.claude/plugins/config/claude-for-legal/product-legal/CLAUDE.md.
Исходный текст
product-legal · Apache-2.0
# Product Counsel Plugin Product legal workflows: launch review, marketing claims review, feature risk assessment, and fast "is this a problem?" triage. Built around a risk calibration learned from your actual launch review history — what blocks at *your* company, not generically. **Every output is a draft for attorney review — cited, flagged, and gated — not a legal conclusion.** The plugin does the work: reads the documents, applies your playbook, finds the issues, drafts the memo. A lawyer reviews, verifies, and decides. Citations are tagged by source so you know which ones came from a research tool and which ones need checking. Privilege markers are applied conservatively so nothing waives by accident. Consequential actions — filing, sending, executing — are gated behind explicit confirmation. ## Who this is for | Role | Primary workflows | |---|---| | **Product counsel** | Launch review, feature risk assessment, calibration maintenance | | **Product managers** | "Is this a problem?" triage self-serve | | **Marketing** | Claims review before ship | | **GC / Legal leadership** | Feature risk assessments for escalated items | ## First run: the cold-start interview Connects to your launch tracker (Jira/Linear), reads ten of your past launch reviews, learns what you actually block vs. what you wave through. Builds a risk calibration table that every other skill reads from. Your configuration is stored at `~/.claude/plugins/config/claude-for-legal/product-legal/CLAUDE.md` and survives plugin updates. ``` /product-legal:cold-start-interview ``` ## Commands | Command | Does | |---|---| | `/product-legal:cold-start-interview` | Cold-start interview | | `/product-legal:launch-review [PRD or ticket]` | Full launch review against your framework | | `/product-legal:marketing-claims-review [copy]` | Marketing claims review | | `/product-legal:is-this-a-problem [question]` | Fast "is this a problem?" answer | | `/product-legal:matter-workspace` | Manage matter workspaces (multi-client private practice only) — new, list, switch, close, none | ## Skills | Skill | Purpose | |---|---| | **cold-start-interview** | Writes ~/.claude/plugins/config/claude-for-legal/product-legal/CLAUDE.md from interview + past launch reviews | | **launch-review** | Category-by-category review, calibrated to your company | | **marketing-claims-review** | Claims taxonomy: puffery/factual/comparative/implied/absolute | | **feature-risk-assessment** | Deep dive on one issue when launch review isn't enough | | **is-this-a-problem** | Same-minute triage for the quick Slack question | | **matter-workspace** | Create, list, switch, and close matter workspaces for multi-client practices; isolates each client/matter so context does not leak across them | ## Interactive commands vs. scheduled agents The commands above run when you invoke them — for when you're working a matter. The agents below run on a schedule — for what moves while you're not looking: | Agent | What it watches | Default cadence | |---|---|---| | **launch-watcher** | Launch tracker (Jira/Linear) for upcoming launches that likely need legal review; filters tickets with launch dates in the next 30 days per the calibration table | Daily | ## Integrations **Connect a research tool first — the citation guardrails depend on it.** Without one, every cite is tagged `[verify]` and the reviewer note above each deliverable records that sources weren't verified. Skills work either way; a research tool (CourtListener) just shifts verification work off your plate. Ships with connectors configured in `.mcp.json`: - **Slack** — search messages, read channels, find discussions (general bucket) - **Google Drive** — search, read, and fetch documents (general bucket) - **Linear** — issue tracking and project management - **Atlassian** — Jira issues and Confluence pages - **Asana** — tasks and project tracking With a tracker connected: cold-start pulls launch history, launch-review pulls ticket context, launch-watcher agent monitors the calendar. ## Quick start ``` /product-legal:cold-start-interview ``` Then: ``` /product-legal:is-this-a-problem "Can we A/B test the pricing page?" ``` → Same-minute answer calibrated to your risk table. ``` /product-legal:launch-review PROJ-1234 ``` → Full review, category-by-category, with action items. ## How it learns Your practice profile at `~/.claude/plugins/config/claude-for-legal/product-legal/CLAUDE.md` isn't static — it improves as you use the plugin. Skills tell you when an output used a default you should tune. You can re-run setup, edit the file directly, or tell a skill to record a new position. ## Notes - The calibration table is the whole thing. If it's wrong, every review is wrong. Re-run setup when your risk posture changes (new regulator, new consent decree, new GC). - `is-this-a-problem` is designed for PMs to self-serve. It answers fast and routes to a real review when it should. - Feature risk assessment is for the 10% of launches that need depth. Most don't — don't generate paperwork. ## Prerequisites Some features reference external integrations (document management, launch trackers, eDiscovery, case management, regulatory feeds). These are not bundled — if you have an MCP server for one of these in your environment, the relevant features will use it. Without one, the plugin falls back to file upload and manual workflows. Run `/product-legal:cold-start-interview --check-integrations` to see what's available in your environment. ## Configuration Your configuration is stored at `~/.claude/plugins/config/claude-for-legal/product-legal/CLAUDE.md` and survives plugin updates — you only run setup once.
© Anthropic · Apache-2.0 · anthropics/claude-for-legal