数据分析

office-collaboration-radar-en

试用

Extract an evidence-linked collaboration status card from chat logs, meeting notes, and project updates. Use only when a user explicitly asks for project progress, confirmed decisions, owner-and-deadline actions, blockers, cross-functional dependencies, human-review items, an executive summary, a radar chart, or stable JSON. Do not trigger merely because the user pasted text. Use this English package only when the user requests English output. It processes user-supplied local files, redacts common PII and credential patterns, and writes only to an explicit output path.

它能做什么

Turn scattered collaboration records into a stable, auditable project status card. Extract only facts supported by the supplied material. Do not produce a full weekly report, judge individual performance, send reminders, or write to external systems.

技能文档

Office Collaboration Radar — English

Turn scattered collaboration records into a stable, auditable project status card. Extract only facts supported by the supplied material. Do not produce a full weekly report, judge individual performance, send reminders, or write to external systems.

Chinese version: Office Collaboration Radar

Capabilities

  • Produce seven fixed modules: overview, progress, decisions, actions, risks and dependencies, cross-functional relationships, and human-review items.
  • Attach evidence snippets of at most 80 characters; use Not provided when no reliable evidence exists.
  • Redact phone numbers, email addresses, government IDs, bank-card numbers, and user-supplied entity names.
  • Flag conflicting owners, deadlines, or statuses for human review.
  • Rank actions as P0/P1/P2, synthesize an executive summary, and generate an SVG collaboration-health radar.
  • Export the structured card to CSV, Feishu/Bitable mapping JSON, or Notion mapping JSON.

Input

Accept Markdown, plain text, exported chats, meeting notes, or project logs. Label the source of each segment when combining materials. Optional metadata includes project name, time range, departments, and a list of known entities to redact.

If the material does not state a project name, time range, owner, deadline, or department, use Not provided. Do not infer or complete it.

Workflow

  1. Separate facts from opinions, questions, tentative statements, small talk, and duplicates.

  2. Draft the seven-module card using the key order in templates/json-output-schema.md.

  3. Keep owner and deadline values verbatim. If an exact value cannot be found in the supplied source, replace it with Not provided.

  4. Enforce evidence, redaction, Markdown-cell sanitization, conflict checks, and priority ordering:

    python3 scripts/process.py enforce --card draft-card.json --source source.txt --out enforced-card.json
    

    Aggregate multiple cards first when needed:

    python3 scripts/process.py aggregate --cards card-1.json card-2.json --out aggregated.json
    
  5. Optionally generate a radar chart:

    python3 scripts/radar_chart.py generate --card enforced-card.json --out collaboration-radar.svg
    
  6. Optionally export a downstream mapping:

    python3 scripts/export_card.py export-card --card enforced-card.json --format csv --out actions.csv
    
  7. Return the Markdown card, JSON result, and executive summary. Before publishing or changing the skill, run:

    python3 scripts/process.py selftest
    python3 scripts/validate_output.py
    

Output contract

Keep every heading in this order, even when a section contains only Not provided:

# Collaboration Status Card

## Project Overview
## Progress
## Confirmed Decisions
## Owner × Deadline Actions
## Risks / Blockers / Dependencies
## Cross-functional Relationships
## Human Review Required
## JSON Output

Use templates/collaboration-status-card.md for the Markdown layout and templates/json-output-schema.md for the machine-readable contract.

Evidence and safety boundaries

  • Do not turn tentative intent into a commitment or invent owners, deadlines, decisions, departments, or launch dates.
  • Use Conflict detected; human review required consistently and add a matching human-review item.
  • Sanitize all user-derived Markdown table values and escape CSV formula prefixes.
  • Escape SVG titles before rendering them.
  • Minimize and redact secrets, tokens, customer data, and sensitive HR information.
  • Do not send messages, email, or calendar invitations, and do not modify Jira, Feishu, Notion, CRM, calendars, or other systems.
  • Do not convert interpersonal friction or emotional language into performance judgments or blame.

The example under examples/enterprise-ai-pilot/ is synthetic and redacted. It demonstrates extraction behavior only; it is not evidence of a production deployment or business outcome.

相关技能

把自然语言描述转为结构化 JSON,并由 mcp-diagram-generator MCP 服务生成 Draw.io、Mermaid 或 Excalidraw 图表文件。

作者 nssa.io1.0k 次安装47 星标

诊断生产力系统反复失效的根因,给出最小干预——容量测算、瓶颈定位、可靠的本地记录。

作者 Iván855 次安装69 星标

以 AI 机器人身份加入视频会议,提供语音、虚拟形象与屏幕共享四种模式。

作者 johnpatternai22 次安装8 星标

在本地磁盘以分类纯 Markdown 文件保存需要长期留存的事实,与智能体内置记忆并存。

作者 Iván559 次安装18 星标

从 AdMapix API 拉取广告创意、应用、榜单和收入预估等数据,原样返回结构化 JSON。

作者 fly0pants4.3k 次安装296 星标

执行 Git 操作(提交、分支、合并、变基、冲突解决与恢复)时强制套用安全规则。

作者 Iván535 次安装31 星标

yamyeed 的更多技能

浏览全部技能

根据身体数据、目标和生理周期,计算每日热量、宏量营养素并生成饮食建议。

作者 yamyeed16 次安装1 星标

Provide English-language everyday nutrition support, including calorie and macro estimates, meal planning, food substitutions, restaurant estimates, and user-requested food or hydration logs. Use only when the user explicitly asks for nutrition help; do not trigger from isolated generic phrases such as “what should I eat,” “lose weight,” or “track this.” Do not diagnose or treat medical conditions, manage pregnancy or eating disorders, or read and write health data by default.

作者 yamyeed2 次安装1 星标

微信读书/WeRead 专用阅读数据与笔记分析 Skill。仅当用户明确提到“微信读书”或“WeRead”,并请求看板、书籍分析、跨书关联、笔记导出或阅读人格分析时使用。 通过用户提供的 WEREAD_API_KEY 读取完成当前任务所需的最小范围数据;每次联网前说明范围并取得确认,写本地文件前另行说明内容和路径并取得确认。

作者 yamyeed1 次安装1 星标

把用户明确提供的中文群聊记录、会议纪要或项目更新整理为证据可追溯的协作状态卡片。 仅在用户明确要求分析项目进展、决策、Owner 与截止时间、风险、阻断、依赖或需确认项时使用。 不因用户仅粘贴文本而自动触发,不发送催办,不写入外部系统;读取和写入本地文件都必须使用用户明确指定的路径。

作者 yamyeed1 次安装

中文阅读、学习与 Obsidian 知识管理助手。仅在用户明确要求使用“书童”,或明确要整理书架/笔记、制作读书笔记、概念讲解、背诵测验、备考计划、模拟面试、学习型写作或维护 Obsidian Vault 时使用。 不因“摘要”“找到”“提问”“润色”等孤立通用词自动触发;不默认联网、落盘、删除文件或创建提醒。

作者 yamyeed15 次安装1 星标