teams
$
npx mdskill add openai/plugins/teamsProcess Teams chats and Planner tasks for actionable follow-ups.
- Extracts owners and next steps from unread conversations.
- Integrates with Microsoft Teams and Planner services.
- Decides actions by matching user requests to workflow paths.
- Delivers drafts, messages, or task updates directly.
SKILL.md
.github/skills/teamsView on GitHub ↗
--- name: teams description: Summarize Microsoft Teams conversations, triage unread or recent activity, draft follow-ups, and manage Planner tasks through connected Teams data. Use when the user wants to review chats or channels, identify owners and next steps, prepare a safe reply or post, or turn Teams follow-ups into Microsoft Planner tasks. --- # Teams ## Overview Use this skill to route Microsoft Teams work into the right workflow: summarize channels, review recent activity, draft replies, send messages, extract follow-ups, or manage Planner tasks from Teams context. Keep answers grounded in exact Teams context, preserve thread intent, and use the write path that matches the user's requested action. ## Related Skills | Workflow | Skill | | --- | --- | | Compose, route, draft, or send Teams messages | [../teams-messages/SKILL.md](../teams-messages/SKILL.md) | | Summarize one Teams channel or scoped conversation | [../teams-channel-summarization/SKILL.md](../teams-channel-summarization/SKILL.md) | | Build a daily digest across selected chats or channels | [../teams-daily-digest/SKILL.md](../teams-daily-digest/SKILL.md) | | Find messages that likely need a response and draft replies | [../teams-reply-drafting/SKILL.md](../teams-reply-drafting/SKILL.md) | | Triage what likely needs the user's attention | [../teams-notification-triage/SKILL.md](../teams-notification-triage/SKILL.md) | | Review, create, update, and delete Microsoft Planner tasks from Teams follow-ups | [../teams-planner-task-management/SKILL.md](../teams-planner-task-management/SKILL.md) | ## Support Checks - Confirm the requested Teams action is supported before collecting extra details. If the connector cannot do it, say so immediately and offer the closest supported path. - Verify write capability before concluding a Teams action is unsupported. Distinguish between: - the needed tool not being surfaced in-session - the connector truly lacking the capability - the destination or Teams product rules blocking the action - For any write request involving DMs, group chats, channels, or replies, resolve the exact destination first. ## Core Truths - Unread state exists for chats only. The connector does not expose unread markers for specific channel messages. - Mention metadata is reliable on chat and channel message-history reads. Do not rely on Teams search hits to detect mentions. - Teams does not expose native persisted drafts here. "Draft" means return draft text when the requested action is drafting rather than posting. - There is no Slack-canvas analogue in this Teams connector. If the user wants something posted in Teams, return or send message text rather than inventing a document workflow. - Real outbound Teams mentions require structured mention inputs with exact Entra user IDs. Do not rely on plain `@name` text. - Planner in this plugin is the Microsoft Planner task surface reached from Teams workflows. Treat it as shared Microsoft task infrastructure, while keeping this plugin focused on Teams-originated follow-ups. - For unbounded channel summaries, start with `list_channel_messages(top=50)`. Do not probe larger values by default because the underlying endpoint rejects oversized reads. - If Teams review produces action items and the user wants tracked work instead of a message draft or private list, route to [../teams-planner-task-management/SKILL.md](../teams-planner-task-management/SKILL.md). ## DM Routing - When the user refers to an existing DM or group chat, prefer resolving that chat instead of creating a new one. - For a new direct chat, resolve the target user first and use `create_chat(chat_type='oneOnOne')` with exactly one recipient user ID. - If one-on-one chat creation fails with a caller-membership or contract mismatch, use the known-good fallback path: create a two-person `group` chat containing the caller and the intended recipient, then send the message there. - For note-to-self requests, prefer an obvious existing self-chat target if one is available. Otherwise use the supported one-member `group` chat fallback. ## Write Safety - Preserve participant names, dates, links, files, decisions, and action items from the source conversation unless the user asks to change them. - Treat channel-wide announcements, broad mentions, and shared-thread edits as high-impact. Call them out before posting. - If multiple chats, channels, or similarly named meetings are in scope, identify the intended destination before drafting or posting. - For answer-in-thread requests, post, send, or reply when the user has clearly asked for that action. - Use canonical message paths for replies whenever possible. Do not treat free-form quoted text as a stable reply target. - If outside context is needed to answer well, use the narrowest extra Teams context that materially changes the answer. - If a write request fails after capability verification, say whether the blocker is connector availability, target resolution, or a Teams product rule such as chat membership requirements. ## Output Conventions - Distinguish clearly between a private summary for the user and a message intended for Teams. - Lead summaries with the latest status, then list decisions, owners, blockers, and next steps. - Keep post-ready drafts concise, with one clear objective and a concrete ask when needed. - When some channel activity is unreadable or artifact-only, say so explicitly instead of presenting it as confirmed human conversation. ## Example Requests - "Summarize the latest Teams thread with design and tell me what follow-ups came out of it." - "Check what likely needs my attention in Teams and separate unread chats from recent channel activity." - "Draft a short Teams reply that confirms the rollout plan and asks for final QA sign-off." - "Turn this Teams meeting follow-up list into Planner tasks in the launch plan." ## Light Fallback If Teams data is missing or incomplete, say that Teams access may be unavailable, pointed at the wrong destination, or too broad to answer reliably, then ask the user to reconnect or narrow the scope.
More from openai/plugins
- accessibility-and-inclusive-visualizationMake data visualizations accessible and inclusive. Use when the user needs chart or diagram accessibility guidance, text alternatives for complex visuals, color and contrast review, keyboard support, reduced-motion behavior for animation or parallax, or an accessibility QA workflow for exported figures, UML-like diagrams, and dashboards.
- agent-browserBrowser automation CLI for AI agents. Use when the user needs to interact with websites, verify dev server output, test web apps, navigate pages, fill forms, click buttons, take screenshots, extract data, or automate any browser task. Also triggers when a dev server starts so you can verify it visually.
- agent-browser-verifyAutomated browser verification for dev servers. Triggers when a dev server starts to run a visual gut-check with agent-browser — verifies the page loads, checks for console errors, validates key UI elements, and reports pass/fail before continuing.
- agents-sdkBuild AI agents on Cloudflare Workers using the Agents SDK. Load when creating stateful agents, durable workflows, real-time WebSocket apps, scheduled tasks, MCP servers, or chat applications. Covers Agent class, state management, callable RPC, Workflows integration, and React hooks. Biases towards retrieval from Cloudflare docs over pre-trained knowledge.
- ai-elementsAI Elements component library guidance — pre-built React components for AI interfaces built on shadcn/ui. Use when building chat UIs, message displays, tool call rendering, streaming responses, reasoning panels, or any AI-native interface with the AI SDK.
- ai-gatewayVercel AI Gateway expert guidance. Use when configuring model routing, provider failover, cost tracking, or managing multiple AI providers through a unified API.
- ai-generation-persistenceAI generation persistence patterns — unique IDs, addressable URLs, database storage, and cost tracking for every LLM generation
- ai-sdkVercel AI SDK expert guidance. Use when building AI-powered features — chat interfaces, text generation, structured output, tool calling, agents, MCP integration, streaming, embeddings, reranking, image generation, or working with any LLM provider.
- aiq-deploy|
- aiq-research|