Minutes Weekly logo

Minutes Weekly

CommunityPopular
silverstein
minutes-weekly

Weekly meeting synthesis — themes, decision arcs, stale commitments, and what deserves your attention next week. Use when the user says "weekly review", "what happened this week", "weekly summary", "recap my week", "any outstanding items", "week in review", or at the end of a work week.

Overview

Publishersilverstein
Repositoryminutes
Skill nameminutes-weekly
Stars
1.5K
Forks
163
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 silverstein on GitHub. Read the source before you install it.

Installation

Install the Minutes Weekly 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.
  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/silverstein/minutes.git /tmp/minutes
mkdir -p .claude/skills
cp -r /tmp/minutes/tooling/skills/goldens/claude/minutes-weekly .claude/skills/minutes-weekly
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Minutes Weekly 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 Minutes Weekly 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 Minutes Weekly 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.

/minutes-weekly

Synthesize a week of policy-authorized normal meetings and voice memos into a forward-looking brief — themes, decision arcs, stale commitments, and what deserves attention Monday. Restricted history is excluded by default.

How it works

This is a synthesis skill, not a command wrapper. It reads across the normal meetings authorized by Minutes, cross-references decisions and action items, and produces an intelligence brief. Restricted meetings are excluded from this agent workflow by default and must never be treated as absent facts.

Phase 1: Gather this week's recordings

bash
minutes list --limit 50

Filter to recordings from the last 7 days by checking the date field in each result's frontmatter. Do NOT use minutes search with a date string as the query — that searches file content, not dates.

For each selected recording, run minutes get "<exact path>" --json, require exit status 0, and use only the returned content. List/search paths are hints, not retained capabilities; never reopen them with the host Read tool.

If zero recordings this week: Say: "No recordings found for the past 7 days. Nothing to synthesize." Offer: "Want me to look at the past 2 weeks instead?" Do NOT hallucinate a weekly summary.

If only 1-2 recordings: Still produce the brief — it's shorter but still valuable. Note: "Light week — only [N] recordings."

Phase 2: Theme extraction

Identify the 3-5 dominant themes across the authorized normal meetings this week. A theme is a topic that appeared in 2+ meetings.

For each theme:

  • Which meetings discussed it
  • How the conversation evolved across meetings
  • Whether a resolution was reached

Present as:

## This Week's Themes

### 1. Pricing (3 meetings)
- Mon: Case proposed $599 baseline
- Wed: Alex pushed back to annual billing
- Fri: Agreed on monthly billing experiment
- Status: RESOLVED (after 3 discussions)

### 2. Q2 Roadmap (2 meetings)
- Tue: Committed to April ship date
- Thu: Discussed pushing to May
- Status: CONFLICTING — not reconciled

Phase 3: Decision evolution arcs

For every decision found in the authorized normal sources this week, search for prior normal-source decisions on the same topic in the last 30 days:

bash
minutes search "<topic>" --since <30-days-ago> --limit 20

Require search exit status 0. Reauthorize any result used in the decision arc with minutes get "<exact path>" --json; never read a result path directly.

Classify each decision:

  • STABLE — Decision held across 2+ mentions → "Locked in"
  • VOLATILE — Changed 2+ times in 14 days → "Still in flux"
  • CONFLICTING — Two active contradictory decisions → "Needs reconciliation"
  • NEW — First time this topic was decided → "Fresh"

Present as a table:

## Decision Arcs

| Decision | Status | Arc | Last Meeting |
|----------|--------|-----|-------------|
| Pricing: monthly billing experiment | VOLATILE | $599→annual billing→monthly billing | Fri w/ Alex |
| Hire senior eng by Q2 | STABLE | Set Mar 10, held | Tue w/ Case |
| Q2 ship date | CONFLICTING | April vs May | Thu w/ team |

If there are CONFLICTING decisions, flag them prominently: "⚠️ You have conflicting decisions on Q2 ship date. Monday is a good time to reconcile."

Phase 4: Action item audit

Scan authorized normal meetings from the past 30 days for action items:

bash
minutes actions

Require exit status 0. If the native action surface fails, report the source unavailable; never grep meeting files as a fallback.

Categorize:

  • Completed this week — items that were marked done
  • Still open, on track — items with future due dates
  • Overdue — items with past due dates, still open
  • Assigned to others — items others owe the user

Flag overdue items prominently: "⚠️ 3 action items are overdue. The oldest is from Mar 10 (pricing doc for Alex)."

Phase 4b: Relationship intelligence

Run minutes people --json and minutes commitments --json for bounded relationship intelligence. Require exit status 0 from each; a projection failure means unavailable, never “no relationships” or “no commitments.” Continue using source-backed actions and consistency results from the surrounding phases.

Phase 5: Unresolved preps

Scan ~/.minutes/preps/ for prep files from this week that were never followed by a debrief:

bash
ls ~/.minutes/preps/ 2>/dev/null

For each prep file from the last 7 days: check if a meeting with that person occurred after the prep date. If the user prepped but never debriefed:

"You prepped for a call with Alex on Tuesday but I don't see a debrief. Did the call happen?"

This catches meetings that happened but weren't recorded, or recordings that weren't processed.

Phase 6: Forward brief

Produce a "what deserves your attention Monday" section:

Before deciding how to order the weekly output, check:

bash
node "${CLAUDE_PLUGIN_ROOT}/hooks/lib/minutes-learn-cli.mjs" get-presentation-focus weekly

If the result is:

  • decisions-first → lead with Decision Arcs and unresolved conflicts before commitments
  • commitments-first → lead with Action Item Audit / stale commitments before decision arcs
  • memo-heavy → surface voice memos and idea capture much more prominently in the synthesis

Use this concrete ordering:

  • decisions-first

    1. Decision Arcs
    2. This Week's Themes
    3. Action Item Audit / stale commitments
    4. Attention Monday
  • commitments-first

    1. Action Item Audit / stale commitments
    2. Decision Arcs
    3. This Week's Themes
    4. Attention Monday
  • memo-heavy

    1. This Week's Themes, but ensure voice memos and idea capture are surfaced explicitly inside the theme summary
    2. Action Item Audit
    3. Decision Arcs
    4. Attention Monday

If there is no preference, keep the default order in this skill.

## Attention Monday

1. **Reconcile Q2 ship date** — April vs May is unresolved.
   Last discussed Thu with the team.

2. **Send pricing doc to Alex** — Overdue since Friday.
   She's expecting it. This is blocking.

3. **Follow up with Case on competitor grid** — He committed Mar 17.
   No update yet. Worth a ping.

Prioritize by: CONFLICTING decisions > overdue action items > open commitments > unresolved preps.

Phase 7: Closing ritual

End with three beats:

  1. Signal reflection — Reference a pattern or insight from the week. "Pricing dominated this week — 3 meetings, 3 changes, one resolution. That's behind you now."

  2. Assignment — The single most important thing to do Monday. "First thing Monday: send the pricing doc to Alex. It's overdue and she's waiting."

  3. Next skill nudge — "For your most important meeting Monday, run /minutes-prep to go in prepared."

Gotchas

  • Record explicit weekly presentation preferences when the user states them. If the user says "always show commitments first", "start with decisions", or "surface voice memos more", persist it:
    bash
    node "${CLAUDE_PLUGIN_ROOT}/hooks/lib/minutes-learn-cli.mjs" set-presentation-focus weekly commitments-first "User explicitly prefers commitments first in weekly synthesis"
    node "${CLAUDE_PLUGIN_ROOT}/hooks/lib/minutes-learn-cli.mjs" set-presentation-focus weekly decisions-first "User explicitly prefers decisions first in weekly synthesis"
    node "${CLAUDE_PLUGIN_ROOT}/hooks/lib/minutes-learn-cli.mjs" set-presentation-focus weekly memo-heavy "User explicitly wants stronger voice-memo emphasis"
  • Zero recordings is not an error — Say "nothing this week" clearly. Don't hallucinate a summary. Offer to extend the range.
  • Light weeks (1-2 recordings) are still worth summarizing — A single meeting can have important decisions. Don't dismiss it.
  • Don't be a nag about overdue items — Surface them factually. "This is overdue since Friday" not "You really need to get on this."
  • Decision evolution spans beyond this week — Search the last 30 days for related decisions, not just this week. A decision made today might conflict with one from 3 weeks ago.
  • Preps without recordings might be intentional — Maybe the meeting was cancelled. Ask, don't assume.
  • Voice memos count — Include ~/meetings/memos/ in the weekly scan. Ideas captured on the go are easy to forget.
  • End-of-week timing — The user might run this Thursday or Monday morning. "This week" should be the most recent 7 days, regardless of day-of-week.

Frequently asked questions

What does the Minutes Weekly AI skill do?

Weekly meeting synthesis — themes, decision arcs, stale commitments, and what deserves your attention next week. Use when the user says "weekly review", "what happened this week", "weekly summary", "recap my week", "any outstanding items", "week in review", or at the end of a work week.

Why use Minutes Weekly on TypingMind?

Because you install it once and use it with any model. Minutes Weekly 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 Minutes Weekly in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/silverstein/minutes/tree/main/tooling/skills/goldens/claude/minutes-weekly. TypingMind reads its SKILL.md and installs it as a skill you can enable per chat.

Which AI models can use Minutes Weekly?

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 Minutes Weekly?

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

Is the Minutes Weekly AI skill free?

Yes. It is published on GitHub by silverstein 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 👇