Hire an AI meeting minutes and action tracker

Everything it can hand back

Your task sets the actual outputs. The specialist picks the methods that cover them.

• Short minutes with decisions, actions and open questions
• Every action from last week marked done, open, dropped or not mentioned
• A list of what is late or has slipped, with the likely reason
• A draft agenda for the next meeting, led by the late items
• A question about the meeting itself, only if it keeps ending with no decisions

How it does the job well

• Turn every request into owned actions: who, what and by when.
• Work from the real calendar, notes and lists. Never invent meetings, people or dates.
• Put the most important decision or deadline first, then the details.
• Prepare drafts and proposals. Sending, booking and changing records waits for approval.

Mistakes it avoids

• Marking an action done because nobody mentioned it. Use "not mentioned" and ask.
• Naming a person as the cause of a slip. Name the reason the notes give.
• Letting the agenda grow longer every week. Drop what is decided.
• Recording private or personal remarks from the room. Minutes record decisions and actions.
• Losing commitments buried in long notes.
• Mixing confirmed facts with guesses about other people's plans.

What a good result looks like

An example, not your task: You run a weekly team or project meeting and actions keep getting lost.

• This week. One line with actions done, still open, dropped and new, plus the path taken: healthy, slipping or not working.
• Minutes. The bottom line, decisions, actions with owner and date, and open questions.
• Late and slipping. Each late action with its owner, weeks open and the likely reason from the notes.
• Questions for the owner. Missing owners or dates, and a go or no-go question on the meeting if it keeps ending with no decisions.
• Next agenda. A draft agenda with late items first, then open questions, then new topics. Not sent.
• Files and missing. The file saved for each step and every MISSING line in one list.

How the work is checked

• Every action has an owner and a date, or is marked [unassigned] or [TBD].
• Every action from last week has a status. None disappeared.
• Nothing was invented that is not in the notes.
• The late items lead the agenda.
• States which checks passed, failed or could not run, and names missing inputs.

These checks have not been run against a model yet. Your AI uses them to check its own work.

Capabilities

• Write the minutes. Turns the notes into a bottom line, decisions, actions and open questions. A missing owner or date is marked, never invented.
• Compare with last week. Sets each old action to done, still open, dropped or not mentioned, adds the new ones and counts weeks open.
• Choose the path. Healthy goes straight to the agenda. Slipping puts the stuck items first and asks what would unblock them. Not working asks whether to hold the meeting.
• Check the meeting itself. Only when items slip three weeks running or two meetings end with no decision, it works out a go or no-go and shows it to you.
• Draft the next agenda. Opens with the late items, then open questions, then new topics. Anything already decided is dropped.

How the role works

1. Checks your task, context and available tools.
2. Picks the methods needed and prepares a work plan.
3. Carries out supported tasks, with each method's limits and review gates.
4. Checks the real deliverables and corrects supported faults.
5. Hands back files, evidence and unresolved gaps.

Your specialist's working setup

Download the base specialist without a task form or account. README.md and SETUP.md explain how to use the folder. SOUL.md defines its scope. INTAKE.md asks missing questions inside your AI. PLAYBOOK.md, SKILLS.md, BEST_PRACTICES.md, CHECKS.md and HANDOFF.md guide the work and its checked deliverables.

Complete associated skills, supporting files and licences are included. Choose only the methods needed for the actual task. Keep context, working files and accepted feedback in your AI project.

Use this specialist through MCP or the API

Call prepare_specialist with role_id weekly-meeting-actions-keeper only. No task is required to load it. Your AI asks what job you need done, what you want back and which relevant context it can use.

For task-led hiring, call find_specialist, explain the suggested specialists and let the user accept or change the team.

Complete ZIP: Download base specialist. API: GET /api/v1/specialists/weekly-meeting-actions-keeper/pack. The host AI supplies its own model, tools and execution; a pack does not connect accounts or run a task.

Inputs and helpful context

The specialist starts from the task brief. Only active methods require their inputs.

For the primary method, when used

• This week's rough notes or transcript, with who was there
• Last week's action list (from the second week on), with a row per action: action, owner, date, status and weeks open
• What the meeting is for (optional, the first time)

Helpful context (optional)

• The meeting's standing agenda
• A list of attendees and their roles

Tools and access

• Claude Code: It can keep the actions file and the folder of past weeks, and run after each meeting.
• Google Meet: Use a transcript from the call as the input. Sending the result stays with you.
• Claude Code or an AI that can run Python: It runs the three scripts, so the sums and the verdict are exact.
• Any chat AI: It works the cost out by hand and still prints the verdict line.

Functions and deliverables

• Meeting minutes and action tracker: Turn meeting notes into minutes and track earlier commitments. Get overdue questions and a next agenda with stuck work first.
  • Short minutes with decisions, actions and open questions
  • Every action from last week marked done, open, dropped or not mentioned
  • A list of what is late or has slipped, with the likely reason
  • A draft agenda for the next meeting, led by the late items
  • A question about the meeting itself, only if it keeps ending with no decisions
• Meeting minutes writer: Turn rough meeting notes into short minutes. You get decisions, action items with owners and dates, and open questions, ready to send.
  • A three-line bottom line
  • A list of decisions
  • Action items with owner and date
  • Open questions with a next step
  Source: sickn33.
• Task list manager: Keep one task list file current. Tell your AI what you add, finish or wait on, and it moves each item to the right section.
  • A TASKS file with four clear sections
  • Items moved and updated as you report
  • A review of open and stuck work
  Source: Anthropic.
• Meeting pre-read writer: Turn meeting goals and connected Notion pages into a timed agenda and linked pre-read.
  • A timed agenda with owners
  • A pre-read with links to source pages
  • A template matched to the meeting type
  • Short research notes with sources (optional)
  Source: OpenAI.
• Meeting agenda and cost planner: Describe the decision, people and timing. Check whether to meet, estimate its cost and build a timed agenda.
  • A one-line verdict, MEET, ASYNC or NOT-READY, with the total cost in dollars
  • The cost per minute of the meeting
  • A timed, decision-first agenda with a five-minute closing slot
  • An action checklist by owner, with missing owners and dates flagged
  Source: Alireza Rezvani.
• me.md builder: Answer a few questions, or let your AI use what it remembers. Get one private file that tells any AI who you are.
  • One me.md file with up to ten dated sections
  • A list of facts taken from memory, for you to check
  • A short note of what it removed for privacy
  • The two next steps, to save it and set it up

Tasks you can use it for

• You run a weekly team or project meeting and actions keep getting lost.
• You want to see which items have slipped for weeks, without reading old notes.
• You want the next agenda ready, led by what is still open.

Limits to check before starting

• You hold a one-off or ad hoc meeting: use a plain minutes template or write a short note yourself, since this agent needs last week's list to compare.
• You only want a running action list with no minutes or agenda: use the Task list manager on its own.
• You want to send the minutes or cancel a meeting automatically: this agent never sends or cancels, so do that step yourself.

Best way to use this role

Give it one clear outcome, the actual files and a definition of done. Keep the work in one AI project so it can reuse checked context and earlier files.

• Start a chat with the agent. In Claude Code, add it as a skill or save it in .claude/agents.
• After the meeting, paste your notes or the transcript and ask it to run this week.
• Read the late list and the questions. Fill in any missing owner or date.
• Send the minutes and agenda yourself, or say yes to one named send.
• Run it again after next week's meeting. It reads the saved action list.

Read it as data

Read the complete role setup as JSON. Connect your AI to the AI Blocks MCP library and read aiblocks://roles/weekly-meeting-actions-keeper for its role instructions and pinned sources.

This is a prepared AI role. No task has run. Your AI carries out authorized work using its available tools and reports the actual deliverables and checks.