Channel Digest logo

Channel Digest

OrganizationPopular
microsoft
channel-digest

Summarize activity across multiple Teams channels into a single consolidated digest — key discussions, decisions, mentions, and action items.

Overview

Publishermicrosoft
Repositorywork-iq
Skill namechannel-digest
Stars
1K
Forks
140
Bundled files
Instructions only
Links
  • Markdown instructions

    A SKILL.md file the model loads on demand, so it only costs tokens when a request actually matches.

  • Works with any LLM

    AI skills are plain Markdown, not provider-specific code, so this works with GPT, Claude, Gemini, Grok, or a local model.

  • Self-contained

    Everything the model needs lives in the instructions — no extra files to sync.

  • Open source

    Published by microsoft on GitHub. Read the source before you install it.

Installation

Install the Channel Digest AI skill in TypingMind to use it with any LLM, or drop it into another agent that reads SKILL.md.

1

Install in TypingMind

TypingMind installs a skill straight from its GitHub folder — it reads SKILL.md, bundles the resource files, and stores the result locally.

  1. Open the app and go to Plugins → Skills.
  2. Choose "Install from GitHub".
  3. Paste the skill folder URL below and confirm.
    https://github.com/microsoft/work-iq/tree/main/plugins/workiq-productivity/skills/channel-digest
  4. Enable the skill in any chat where you want it available.
Plugins → Skills → Add skill → From GitHub URL, then paste the folder URL and press Continue.
2

Install in another agent

Any agent that reads the Agent Skills format can use this skill — copy the folder into that agent's skills directory.

Claude Code — .claude/skills
git clone --depth 1 https://github.com/microsoft/work-iq.git /tmp/work-iq
mkdir -p .claude/skills
cp -r /tmp/work-iq/plugins/workiq-productivity/skills/channel-digest .claude/skills/channel-digest
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Channel Digest in any TypingMind chat and the model takes it from there. Its name and description sit in the system prompt, and the moment a request matches, the model loads the full instructions itself — you never invoke it by hand, and it costs no tokens until it is actually used.

The model loads Channel Digest on its own as soon as a request matches it.

Works with any AI model

AI skills are plain Markdown instructions rather than provider-specific code, so Channel Digest is not tied to the model it was written for. Install it once in TypingMind and use it with GPT-5, Claude, Gemini, Grok, DeepSeek, Mistral, Llama, or a local model you run yourself — all on your own API keys.

  • Loaded only when it is needed

    The system prompt carries just the name and description. The instructions are fetched on the first matching request, so an idle skill costs nothing.

  • Switch models mid-chat

    Because the skill is instructions rather than code, changing model does not break it — the next model reads the same SKILL.md.

Skill instructions

This is the SKILL.md content the model loads. Read it before installing — a skill is instructions your model will follow.

Channel Digest

Produce a consolidated summary of activity across multiple Microsoft Teams channels. Scans recent messages from each channel, identifies key discussions, decisions made, action items, and @‑mentions — then presents everything in a single, scannable digest. Ideal for managers, leads, or anyone tracking multiple project channels.

When to Use

  • "What happened in my channels today?"
  • "Summarize the last 3 days across all my project channels"
  • "Give me a digest of Engineering and Product channels"
  • "Any decisions made in the team channels this week?"
  • "Channel digest for the sprint channels"
  • "What did I miss in the channels while I was out?"

Instructions

Step 1: Identify the User

workiq-ask (
  question: "What is my profile information including display name, email, and time zone?"
)

Extract displayName, email, and timeZone.

Step 2: Discover Teams and Channels

List all teams and their channels:

workiq-ask (
  question: "List all Microsoft Teams teams I belong to and their channels."
)

If the user specified particular channels (e.g., "Engineering #general and #design-reviews"), filter to only those. Otherwise, include all non‑archived channels across relevant teams.

Build a channel inventory:

  • Team name → Channel name
  • Skip channels named "General" in low‑priority teams unless the user explicitly includes them.

Step 3: Pull Recent Messages from Each Channel

For each target channel, fetch recent messages within the lookback window:

workiq-ask (
  question: "Show me the last 30 messages in the '<channel name>' channel of the '<team name>' team including replies. For each message include the sender name, date and time, message content, reply count, any @mentions, reactions count, and any attachments or shared links."
)

For each message, capture:

  • Sender display name
  • Timestamp
  • Content (text body for analysis)
  • Reply count — indicates discussion depth
  • Mentions — anyone @‑mentioned, especially the current user
  • Reactions — high reaction count signals importance
  • Attachments — files or links shared

Step 4: Analyze and Categorize Messages

For each channel, classify the messages into categories:

🔑 Key Decisions — messages containing:

  • Decision language: "decided", "approved", "agreed", "going with", "finalized", "confirmed"
  • Follow‑up with context about what was decided

💬 Active Discussions — threads with:

  • 3+ replies indicating substantive conversation
  • Multiple participants contributing
  • Ongoing debate or open questions

📌 Action Items — messages containing:

  • Task language: "action item", "TODO", "follow up", "need to", "can you", "please"
  • Assignments: "@person please…", "owned by…"
  • Deadlines: "by Friday", "due date", "EOD"

📢 Announcements — messages that are:

  • From team owners or managers
  • Posted with high importance
  • Information‑sharing without expecting discussion

📎 Shared Resources — messages with:

  • File attachments (documents, spreadsheets)
  • Links to external resources, PRs, or documents

👤 Your Mentions — messages where the current user was @‑mentioned or referenced by name.

Step 5: Search for User Mentions Across Channels

Cast a wider net for the user's mentions:

workiq-ask (
  question: "Find all Teams channel messages that mention me or are directed at me in the last <lookback period>. Include the team name, channel name, sender, date, and message content for each."
)

Cross‑reference with messages already captured to avoid duplicates. Add any new mentions to the Your Mentions category.

Step 6: Compile the Consolidated Digest

Organize findings by channel, then surface cross‑channel themes at the top.

Output Format

📰 CHANNEL DIGEST
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
📅 Period: {start date} → {end date}
🔍 Channels scanned: {N}  ·  💬 Messages analyzed: {N}

🎯 CROSS‑CHANNEL HIGHLIGHTS
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
🔑 {N} decisions made  ·  📌 {N} action items  ·  👤 {N} times you were mentioned

Top Decisions:
  1. {Team} › #{Channel}: "{decision summary}" — {Person}, {date}
  2. {Team} › #{Channel}: "{decision summary}" — {Person}, {date}

Action Items Needing You:
  • {Person} in #{Channel}: "{action item}" — due {date}
  • {Person} in #{Channel}: "{action item}"

━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
📢 {Team Name} › #{Channel Name}
   💬 {N} messages  ·  🧵 {N} threads  ·  👥 {N} participants
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━

   🔑 DECISIONS
   • "{Decision summary}" — {Person}, {date}

   💬 KEY DISCUSSIONS
   • "{Topic summary}" — {N} replies, {participants}
     └─ Status: {resolved / ongoing / needs input}

   📌 ACTION ITEMS
   • {Person}: "{task description}" — {due date or "no deadline"}

   📢 ANNOUNCEMENTS
   • {Person}: "{announcement summary}" — {date}

   📎 SHARED FILES
   • {filename} — shared by {Person}, {date}

   👤 YOUR MENTIONS
   • {Person}: "@you {message preview}" — {date}

━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
📢 {Team Name} › #{Channel Name}
   💬 {N} messages  ·  ⏸️ Low activity
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
   • {brief summary of the few messages}

🔇 QUIET CHANNELS (no messages in period)
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
   • {Team} › #{Channel}
   • {Team} › #{Channel}

Parameters

ParameterRequiredDefaultDescription
LookbackNo24 hoursTime window to scan (e.g., "3 days", "this week")
TeamsNoAll user's teamsSpecific teams to include
ChannelsNoAll channelsSpecific channels to include
FocusNoAll categoriesFilter to "decisions", "action-items", or "mentions"
Include QuietNotrueWhether to list channels with no activity
Max MessagesNo30 per channelMaximum messages to pull per channel

Required MCP Tools

MCP ServerToolPurpose
workiq (Local WorkIQ CLI)askUser identity, team/channel discovery, message retrieval, and mention search

Tips

  • Say "digest for Engineering channels only" to narrow the scope and speed up results.
  • Use "channel digest for the last week" before a Monday planning meeting to see what happened while you were out.
  • Use action-item-extractor to pull action items from meeting content after reviewing channel activity.
  • For recurring digests, establish a cadence: "give me my channel digest every morning."
  • Ask "expand on the discussion in #{Channel}" to drill into a specific conversation with full context.

Examples

Example 1 — Morning standup prep (default 24-hour window)

"Give me a channel digest for today."

Scans all of the user's Teams channels for the past 24 hours, surfaces any decisions made overnight, flags action items assigned to the user, and lists @-mentions — all in a single digest organized by team and channel.


Example 2 — Catch up after time off (multi-day, scoped channels)

"Channel digest for the Engineering and Product channels over the last 5 days."

Limits the scan to channels in the Engineering and Product teams, pulls up to 30 messages per channel across the 5-day window, and highlights key decisions (e.g., architecture choices, scope changes), active threads, and any messages that mentioned the user while they were away.


Example 3 — Decision-focused digest before a planning meeting

"Show me only decisions and action items from the sprint channels this week."

Uses the Focus parameter to filter output to Key Decisions and Action Items only, skipping announcements and low-activity channel summaries. Ideal for quickly building a pre-meeting agenda from what was agreed or assigned across sprint-related channels.

Error Handling

No teams or channels found

  • ask returns an empty result if the user is not a member of any team, or if the account lacks access to the Teams API.
  • Resolution: Confirm the user's account has Microsoft Teams access and is a member of at least one team. Ask the user to specify a team name manually if discovery fails.

Channel message fetch returns empty or partial results

  • ask may return fewer messages than expected if the channel has low activity or if the lookback window predates available message history (Teams message retention policies vary).
  • Resolution: Proceed with available messages and note in the digest that history may be limited. Surface "No messages in period" for those channels in the Quiet Channels section rather than failing silently.

Permission denied on a channel

  • Private channels require explicit membership; ask will indicate an access error if the user is not a member of a private channel even within a team they belong to.
  • Resolution: Skip the inaccessible channel, log it as "access restricted" in the digest output, and continue processing remaining channels.

Search for mentions returns no results

  • The mentions query via ask may return empty results for broad queries or large tenants.
  • Resolution: Fall back to the mentions already captured during the per-channel message retrieval. Note that the mention search could not be completed so the Your Mentions section may be incomplete.

Slow responses with many channels

  • Querying messages from a large number of channels requires multiple ask calls and may be slow.
  • Resolution: Process channels in batches. Inform the user of the delay and suggest narrowing scope with the Teams or Channels parameters.

Frequently asked questions

What does the Channel Digest AI skill do?

Summarize activity across multiple Teams channels into a single consolidated digest — key discussions, decisions, mentions, and action items.

Why use Channel Digest on TypingMind?

Because you install it once and use it with any model. Channel Digest is plain Markdown rather than provider-specific code, so the same skill runs on GPT-5, Claude, Gemini, Grok, or a local model — and you can switch model mid-chat without it breaking. TypingMind runs on your own API keys, so you pay providers directly instead of a per-seat subscription, and your skills and chats stay in your own storage.

How do I install Channel Digest in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/microsoft/work-iq/tree/main/plugins/workiq-productivity/skills/channel-digest. TypingMind reads its SKILL.md and installs it as a skill you can enable per chat.

Which AI models can use Channel Digest?

Any model you connect in TypingMind. AI skills are plain Markdown instructions rather than provider-specific code, so GPT, Claude, Gemini, Grok, and local models can all load this skill when a request matches it.

How many AI models can I use with Channel Digest?

As many as you like. As long as a model supports skills, you can use Channel Digest with it — GPT, Claude, Gemini, Grok, DeepSeek, Mistral, Llama and more — all on TypingMind with your own API keys.

Is the Channel Digest AI skill free?

It is published on GitHub by microsoft. Check the repository for licensing terms. You only pay your own AI provider for the tokens you use.

What are AI skills?

An AI skill is a reusable instruction bundle that teaches an AI model how to do one specific task. It follows the open Agent Skills format: a SKILL.md file with a name and description, plus any scripts, templates or reference files the model may need. The model reads the instructions only when your request matches the skill, so an installed skill costs nothing until it is used.

How are AI skills different from plugins or MCP servers?

A plugin or MCP server gives a model new tools to call — code that runs somewhere and returns a result. An AI skill gives the model knowledge and process instead: how to approach a task, which steps to follow, what good output looks like. Skills are plain Markdown, so they need no server, no API key and no runtime, and they work with any model.

View all

Set up your own AI workspace now

Get notified about new features and future giveaways by subscribing to our newsletter 👇