Meeting Follow Up Pipeline logo

Meeting Follow Up Pipeline

Organization
zapier
meeting-follow-up-pipeline

Turn a completed meeting into a debrief, decisions, action items, ready-to-send follow-up drafts, and proposed writebacks. Use when the user asks to debrief a meeting, process their last meeting, review meeting follow-ups, or run the meeting follow-up pipeline.

Overview

Publisherzapier
Repositorywade-skills
Skill namemeeting-follow-up-pipeline
Stars
99
Forks
10
Bundled files
Instructions only
LicenseMIT
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 zapier on GitHub. Read the source before you install it.

Installation

Install the Meeting Follow Up Pipeline 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/zapier/wade-skills/tree/main/skills/meeting-follow-up-pipeline
  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/zapier/wade-skills.git /tmp/wade-skills
mkdir -p .claude/skills
cp -r /tmp/wade-skills/skills/meeting-follow-up-pipeline .claude/skills/meeting-follow-up-pipeline
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Meeting Follow Up Pipeline 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 Meeting Follow Up Pipeline 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 Meeting Follow Up Pipeline 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.

Meeting Follow-Up Pipeline

Close the loop after a meeting. Turn a finished conversation into clarity and shipped follow-ups, so nothing important gets dropped.

The job in one line:

meeting happened → clarity shipped → nothing important gets dropped

This is not a transcript dump and not a generic recap.

Setup

Fill in the tools you use (or let the agent ask). The skill degrades gracefully — if a source isn't available, it flags that and asks for pasted input instead.

  • [MEETING NOTES TOOL] — where your notes/transcript live (Granola, Fathom, Otter, Fireflies, Fellow, or pasted text)
  • [EMAIL] — your email, for drafting external follow-ups
  • [CHAT] — your team chat, for drafting internal follow-ups (Slack, Teams, etc.)
  • [TASK SYSTEM] — where action items get promoted (Coda, Notion, Asana, Linear, a spreadsheet)
  • [DRAFTS AREA] — where sendable drafts get staged for review before sending (a drafts folder, an "outbox" doc, etc.)

When to Use

  • "Run the meeting follow-up pipeline"
  • "Debrief my last meeting"
  • "Process this meeting"
  • "What are the key follow-ups from that meeting?"

Right after a meeting, during an end-of-day wrap to catch missed follow-ups, or before the next meeting in a chain to avoid dropping context.

Scope Rule: All Meetings Eligible, Not All Get the Same Treatment

Classify the meeting and route it to one of three levels. Run the workflow for all three — just don't force heavy outputs when the meeting doesn't justify them.

Level 1 — Full pipeline: external, customer/prospect, partner, candidate interviews, board/investor, high-stakes internal decision meetings.

Level 2 — Internal substantive: leadership reviews, initiative reviews, staffing/org discussions, important cross-functional working sessions.

Level 3 — Lightweight capture: routine status meetings, normal recurring syncs, low-novelty internal meetings.

Default Target

If the user only says "run the pipeline" / "process this meeting" / "debrief my last meeting":

  1. Use the most recent completed meeting as the default.
  2. If ambiguous, ask the user to pick from the top 3 recent candidates.

Never guess across multiple adjacent meetings if the target is unclear.

Source Priority

  1. [MEETING NOTES TOOL] — notes and transcript (primary)
  2. Calendar — meeting metadata and attendees
  3. [EMAIL] — only when needed for follow-up drafting or relationship context
  4. Reference context — only when a factual claim needs to be verified

If the notes tool is unavailable, say so explicitly and ask the user for pasted notes or bullets.

Core Flow

Step 1: Identify and Classify the Meeting

Determine: meeting name, time, participants, meeting type, and treatment level (1/2/3).

Step 2: Generate the Debrief

Extract:

  • Bottom line (1-2 sentences on what mattered)
  • Decisions (only actual decisions — if none, say so)
  • Action items (owners and due timing when possible)
  • Risks / open loops (what's still unclear or risky)

Keep the debrief under ~250 words before drafts. Don't bury the takeaway in prose.

Step 3: Draft Follow-Ups Inline (Only When Earned)

Good reasons to draft:

  • An explicit promise or next step was made
  • External relationship maintenance
  • Coordination that will fail if not written down
  • Ambiguity that should be closed quickly

Do not force follow-up drafts for low-value internal chatter.

Drafting rules:

  • Direct, specific, no fluff. Match the user's voice.
  • One clear ask per message.
  • Include a concrete next step and owner.
  • Avoid "just checking in" filler.
  • For external recipients, keep it warm but crisp.

Step 4: Propose Promotion Actions (Propose, Don't Auto-Write)

Propose — do not automatically execute:

  • [TASK SYSTEM] additions for important user-owned commitments
  • [DRAFTS AREA] staging when a draft is clearly worth sending later
  • Decision-log entries for meaningful, durable decisions
  • Initiative/project-tracker escalation for repeated or material blockers

Always propose promotion for: user-owned commitments, externally visible promises, due-soon items, repeated open loops, and decisions meaningful enough to matter later.

Usually don't promote: low-signal discussion points, conversational debris, generic chatter, speculative comments with no owner.

Step 5: Hand Off or Stop

End by asking the smallest useful next question:

  • Stage these to [DRAFTS AREA]?
  • Send now?
  • Promote these owned items to [TASK SYSTEM]?
  • Leave as debrief only?

Do not default to staging or sending. Show drafts inline first.

Required Output Shape

markdown
## Source Status
| Source | Status | Notes |
|--------|--------|-------|
| [MEETING NOTES TOOL] | fresh / missing | Notes or transcript pulled |
| Calendar | fresh / missing | Meeting metadata checked |
| [EMAIL] | fresh / not_needed / missing | Used only if follow-up drafting needed |
| Reference context | not_needed / fresh / missing | Only if a factual claim was verified |

## Meeting Type
**[Meeting type] — Level [1/2/3]**

## Bottom Line
[1-2 sentences]

## Decisions
- [Decision]
- [Decision]

## Action Items
| Owner | Task | Due | Notes |
|-------|------|-----|-------|
| [You] | [Task] | [When] | [Context] |
| [Other] | [Task] | [When] | [Context] |

## Draft Follow-Ups
- **To:** [Person] · **Channel:** [Email/Chat]
  - Draft: "[ready-to-send text]"

## Promotion Actions
- **[TASK SYSTEM] proposal:** [What should be added, and why]
- **[DRAFTS AREA] proposal:** [What should be staged, and why]
- **Decision-log proposal:** [If needed]

## Recommended Next Step
- [Single highest-leverage next move]

Edge Cases

  • Multiple adjacent meetings: ask which one to process, then proceed.
  • No notes yet: use the transcript if available; otherwise ask for 3 bullets.
  • Action-item ambiguity: mark as "Needs owner confirmation" instead of guessing.

Anti-Patterns

  • Don't dump transcript text.
  • Don't force a follow-up draft for every meeting.
  • Don't auto-write to your task system without asking.
  • Don't auto-stage or auto-send drafts without asking.
  • Don't promote low-value chatter into durable systems.
  • Don't treat a routine status sync like a board meeting.

Working Standard

After this workflow, the user should know:

  1. What happened
  2. What was actually decided
  3. Who owes what
  4. What message should be sent, if any
  5. What deserves promotion into the execution system

Frequently asked questions

What does the Meeting Follow Up Pipeline AI skill do?

Turn a completed meeting into a debrief, decisions, action items, ready-to-send follow-up drafts, and proposed writebacks. Use when the user asks to debrief a meeting, process their last meeting, review meeting follow-ups, or run the meeting follow-up pipeline.

Why use Meeting Follow Up Pipeline on TypingMind?

Because you install it once and use it with any model. Meeting Follow Up Pipeline 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 Meeting Follow Up Pipeline in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/zapier/wade-skills/tree/main/skills/meeting-follow-up-pipeline. TypingMind reads its SKILL.md and installs it as a skill you can enable per chat.

Which AI models can use Meeting Follow Up Pipeline?

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 Meeting Follow Up Pipeline?

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

Is the Meeting Follow Up Pipeline AI skill free?

Yes. It is published on GitHub by zapier under the MIT license. 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 👇