Agent skill

Minutes Brief

by silverstein in silverstein/minutes

Fast non-interactive briefing before any meeting — auto-detects your next calendar event, pulls relationship history, surfaces open commitments, and produces a one-page brief in under 30 seconds.

MITAuto-check passed

Install Minutes Brief

skills CLI
$ npx skills add silverstein/minutes --skill minutes-brief -a claude-code

Project install by default; add -g for ~/.claude/skills/.

GitHub CLI
$ gh skill install silverstein/minutes minutes-brief --agent claude-code

Project scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).

Manual copy
$ git clone --depth 1 https://github.com/silverstein/minutes.git skills-src && mkdir -p .claude/skills && cp -r skills-src/.opencode/skills/minutes-brief .claude/skills/minutes-brief && rm -rf skills-src

Use ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.

Claude Code skills documentation · loads skills from .claude/skills/

Facts

Skill name
minutes-brief
GitHub stars
1.5k
Token cost
~3.1k tokens
SKILL.md length
1,345 words
Files
1
Skills in repo
24
Repo updated
First seen
Licence
MIT

At a glance

Fast non-interactive briefing before any meeting — auto-detects your next calendar event, pulls relationship history, surfaces open commitments, and produces a one-page brief in under 30 seconds.

  • Works in 6 steps: Determine the target → Gather data in parallel → Read what you actually need → …
  • Give me a quick brief
  • SKILL.md covers Skill Path, How it works and Gotchas
  • Calls node, git and osascript

What it does

Minutes Brief is an agent skill from silverstein/minutes. Fast non-interactive briefing before any meeting — auto-detects your next calendar event, pulls relationship history, surfaces open commitments, and produces a one-page brief in under 30 seconds. Use this whenever the user says "brief me", "give me a quick brief", "what's coming up", "background on my next call", "who am I meeting next", "brief me on Sarah", "I have a call in 10 min", "quick rundown", or right before walking into a meeting. Different from /minutes-prep — brief is the fast hook-fireable version…

Its SKILL.md is about 3.1k tokens, which your agent loads only when the skill is triggered. It is a single SKILL.md file with no bundled scripts. Compatibility notes: opencode

The repository describes itself as: Open-source, local-first Granola/Otter alternative that Claude Code, Codex, Cursor, and any MCP client can query. Meetings, calls, and voice memos transcribed on-device into… The licence is MIT.

When your agent uses it

  • Give me a quick brief
  • Whats coming up
  • Background on my next call
  • Who am I meeting next

Example prompts

  • “brief me”
  • “give me a quick brief”
  • “s coming up”
  • “/minutes-brief”

Requirements

  • Compatibility (from SKILL.md): opencode

Workflow steps

6 steps, taken from the step headings in SKILL.md.

  1. Determine the target
  2. Gather data in parallel
  3. Read what you actually need
  4. Synthesize the brief
  5. Save and display
  6. One-line nudge

What it can do on your machine

Read from SKILL.md and the folder at commit d3285c7. It shows what the files ask for, not the result of running them.

  • Tool permissions

    Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

    From allowed-tools in the SKILL.md frontmatter.

  • Runs code

    Shell commands in SKILL.md call:

    • node
    • git
    • osascript

    From the folder's file list and the shell code blocks in SKILL.md.

  • Network

    No URLs in SKILL.md. Its commands use git, which can reach the network depending on how they are called.

    From URLs in SKILL.md, links to its own repository left out.

  • Credentials

    Names no API keys, tokens, secrets or passwords.

    From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

  • Compatibility

    opencode

    From compatibility in the SKILL.md frontmatter.

Context cost

Minutes Brief loads about 3.1k tokens when it runs. Until then it costs about 168 tokens; SKILL.md has 1,345 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~168
When it runs · the whole SKILL.md, loaded when a task matches
~3.1k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Auto-check passed

The automated check found no risky patterns in SKILL.md.

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.

SKILL.md

The full file from silverstein/minutes at commit d3285c7, republished under its MIT licence (© silverstein). 1,345 words, ~3,098 tokens.

Download SKILL.mdSave it as .claude/skills/minutes-brief/SKILL.md (or your agent's skills folder).
name
minutes-brief
description
Fast non-interactive briefing before any meeting — auto-detects your next calendar event, pulls relationship history, surfaces open commitments, and produces a one-page brief in under 30 seconds. Use this whenever the user says "brief me", "give me a quick brief", "what's coming up", "background on my next call", "who am I meeting next", "brief me on Sarah", "I have a call in 10 min", "quick rundown", or right before walking into a meeting. Different from /minutes-prep — brief is the fast hook-fireable version that doesn't ask questions and doesn't set goals. Use brief when speed matters; use prep when the user wants to think hard about goals first.
compatibility
opencode

Skill Path

Before running helper scripts or opening bundled references, set:

bash
export MINUTES_SKILLS_ROOT="$(git rev-parse --show-toplevel)/.opencode/skills"
export MINUTES_SKILL_ROOT="$MINUTES_SKILLS_ROOT/minutes-brief"

/minutes-brief

Fast, non-interactive briefing that synthesizes your relationship history with someone into a one-page brief — designed so you can read it in 60 seconds before walking into a call.

This is the proactive layer. It's built to be invoked silently by a hook (e.g., 15 min before a calendar event) and to also work as a manual /minutes-brief command. Unlike /minutes-prep, brief asks no questions and sets no goals — it just hands you the facts.

How it works

Phase 0: Determine the target

Three ways the user can invoke this:

1. With a name (/minutes-brief sarah, "brief me on Alex") → Use that name directly. Before searching, check for learned aliases:

bash
node "$MINUTES_SKILLS_ROOT/_runtime/hooks/lib/minutes-learn-cli.mjs" aliases "<name>" 2>/dev/null

If aliases exist, search across all returned variants and treat them as the same person for the rest of the flow. If the user explicitly says two names are the same person, persist it:

bash
node "$MINUTES_SKILLS_ROOT/_runtime/hooks/lib/minutes-learn-cli.mjs" set-alias "Sarah Chen" "Sarah" "User confirmed these refer to the same person"

Then skip to Phase 1.

2. With "auto" or no argument (/minutes-brief, "brief me on my next call") → Auto-detect the next upcoming calendar event. Try sources in order — use the first that works:

  • Google Calendar MCP (best — Claude users):

    gcal_list_events(timeMin: "<now ISO>", timeMax: "<+2hr ISO>", condenseEventDetails: false)

    Filter to events with 2+ attendees, skip all-day events, pick the soonest. Pull attendee names from the event.

  • gog CLI (if installed):

    bash
    gog calendar list --today --json 2>/dev/null
  • Apple Calendar via osascript (every Mac, zero install):

    bash
    osascript -e 'tell application "Calendar" to get {summary, start date} of (every event of every calendar whose start date >= (current date) and start date < ((current date) + 2 * hours))'

If none return anything, ask once: "I can't find an upcoming meeting. Who do you want a brief on?" Then take whatever they say and move on.

3. Hook-fireable mode (/minutes-brief --auto, or invoked silently from a hook) → Same as auto, but never ask questions. If no upcoming meeting and no name, exit silently with no output. Hooks should never spam the user.

Phase 1: Gather data in parallel

The Minutes CLI already does the hard work. Person profiles, search, relationship rankings, and commitments are bounded policy-safe projections. Run the commands below in parallel and require exit status 0 before interpreting their output.

bash
# 1. Bounded live-source person profile.
minutes person "<name>"

# 2. Recent decisions involving them, last 30 days.
#    Clean JSON on stdout by default (insights does NOT accept --json — output is already JSON).
minutes insights --participant "<name>" --kind decision --since <30-days-ago>

# 3. Recent meetings with them, last 60 days. Newline-delimited JSON (one object per line).
minutes search "<name>" --limit 10 --since <60-days-ago> --format json

CLI stream-handling notes — the Minutes CLI is actively developed and its stream contract is not fully settled. Today (0.8.0):

  • minutes person, minutes insights, and minutes search --format json are bounded live-source surfaces.
  • minutes people, minutes commitments, and minutes person fail closed on authorization, resource-budget, or correction races. Never suppress a nonzero exit or translate it into an empty fact.
  • Any nonzero exit from a live-source command means the source is unavailable. Do not interpret empty stdout as “no history.”
  • Do not invent new flags on top of what's shown above — e.g. minutes insights --json is not a real flag, minutes export --since is not a real flag. The CLI will reject unknown flags with a usage error.

If a future CLI release changes any of these contracts, update this skill in the same PR that ships the CLI change.

Before declaring "first meeting on record", verify it with live-source search:

bash
minutes search "<name>" --limit 1 --format json

For multi-attendee meetings, focus on the requested person. Mention the others briefly at the end. Don't try to brief five people at once.

Phase 2: Read what you actually need

Extract the file paths of the most recent 1–2 meetings from a successful minutes search response. Reauthorize and retrieve each meeting through the native bounded surface:

bash
minutes get "<exact path>" --json

Require exit status 0 and use only the returned content. Never pass the path to the host Read tool: search results are hints, not retained authorization capabilities.

Don't read more than two — past three meetings is enough context for a brief, and the brief is supposed to fit on one screen.

Phase 3: Synthesize the brief

Produce a brief in this exact shape — every section is one tight chunk, total fits on one screen:

markdown
# Brief: <Person Name> · <today's date>

**Last conversation** (<date of most recent meeting>): <2–3 sentences in prose, including the emotional tone if discernible from the transcript>

**They've been thinking about**: <comma-separated list of 3–5 hot topics from the last 30 days, ordered by recency × frequency>

**Commitments**: Use `minutes commitments --person "<name>" --json`; require exit status 0.

**Where things stand**: <one-line read of the relationship vibe — warming, cooling, stable, urgent, drifting>

**Open with**: "<a concrete first sentence the user could literally say at the top of the call — references something specific from the last conversation>"

Concrete example of a good "Open with" line:

Bad: "Hey Sarah, how's everything going with Q2?" — generic, could be said to anyone. Bad: "Hi Sarah, hope you're doing well!" — pure filler. Good: "Hey Sarah — last time you mentioned the Q2 hiring freeze was making your roadmap impossible. Did legal come back on the contractor question?"

The good version is specific (cites Q2 freeze, the contractor angle), references something only THIS conversation history would surface, and gives Sarah an immediate hook to pick up the thread.

Rules for the brief:

  • No filler. Cut anything that doesn't change what the user does in the next 60 minutes.
  • Be specific. "They want to talk about Q2" is useless. "Sarah has raised the Q2 hiring freeze in 3 of her last 4 meetings" is useful.
  • Honest about gaps. If minutes search returns nothing, say that no matching meeting history was found. Never infer relationship facts from the unavailable graph.
  • The opening line matters most. This is the punchline. Make it usable verbatim. The user should be able to literally say it.
Show full SKILL.md (561 more words)Show less
Phase 4: Save and display

Save the brief to ~/.minutes/briefs/ so /minutes-debrief can compare against it later:

bash
mkdir -p ~/.minutes/briefs
chmod 700 ~/.minutes/briefs

Write to ~/.minutes/briefs/YYYY-MM-DD-{person-first-name-lowercase}.brief.md with frontmatter:

yaml
---
person: <Full Name>
date: <today ISO>
brief_type: auto | manual
meeting_count: <count from minutes search results used>
trigger: "calendar:<event title>" | "manual"
---

…followed by the brief body. Then chmod 600 the file — briefs contain relationship intelligence and should be private.

Display the brief inline in the response. Don't paraphrase it or summarize it after — let the brief speak for itself.

Phase 5: One-line nudge

End with exactly one line — this skill is about speed:

"Want to think harder about goals? Run /minutes-prep <name>. After the call, run /minutes-debrief."

That's it. No follow-up questions, no "anything else I can help with?". Brief is fast on purpose.

Gotchas

  • Record explicit workflow preferences when the user states them. If the user says something like "default to prep", "always brief me first", or "stop reminding me about meeting prep", persist it:
bash
node "$MINUTES_SKILLS_ROOT/_runtime/hooks/lib/minutes-learn-cli.mjs" set-explicit workflow_preference meeting_prep_mode prep "User explicitly prefers prep"
node "$MINUTES_SKILLS_ROOT/_runtime/hooks/lib/minutes-learn-cli.mjs" set-explicit workflow_preference meeting_prep_mode brief "User explicitly prefers brief"
node "$MINUTES_SKILLS_ROOT/_runtime/hooks/lib/minutes-learn-cli.mjs" set-explicit nudge_feedback meeting_prep_nudge suppress "User explicitly asked to suppress meeting prep nudges"
  • Record explicit person aliases when the user confirms them. If the user says "Sarah and Sarah Chen are the same person" or "Dan means Dan Benamoz here", persist it:
bash
node "$MINUTES_SKILLS_ROOT/_runtime/hooks/lib/minutes-learn-cli.mjs" set-alias "Sarah Chen" "Sarah" "User confirmed alias"
  • Hook-fireable mode is silent on failure. When invoked with --auto and nothing matches, exit cleanly with zero output. Hooks should never spam the user. Errors only matter when the user is actively asking.
  • First-name slugs match the rest of the plugin. Save files as sarah.brief.md, not sarah-chen.brief.md. Slug rules: lowercase, ASCII-only (strip diacritics via Unicode NFKD), replace spaces and punctuation with hyphens, take only the first name token. "María José Pérez" → maria. "Jean-Claude" → jean-claude. Single-name people (Madonna) → madonna. This matches /minutes-prep and /minutes-debrief so they can find each other.
  • If the user types two names, ask once. "brief me on Sarah and Alex" — pick one as the focus and tell the user. "I'll focus the brief on Sarah since you have more history with her, and mention Alex briefly." Don't refuse, don't try to brief both fully.
  • Don't duplicate prep. Brief sets no goals, asks no questions, and never produces talking points. If the user wants any of that, the closing nudge points them at /minutes-prep. Two skills, two jobs — keep the line clean.
  • One screen of output. If the brief body is longer than ~25 lines, you've added filler. Cut it.
  • Multi-attendee meetings focus on one person. Pick the single most-mentioned attendee (highest meeting count) and brief on them. Mention others in one sentence at the end: "Also in the room: Logan, Kim." Don't try to brief everyone — the signal dilutes to noise.
  • Zero history is not an error — but verify it before claiming it. If minutes search returns empty, say that no matching meeting history was found and use only the calendar event. Never invent history.
  • Calendar attendee names are messy. The event might say "Sarah Chen sarah@acme.com" but transcripts say "Sarah" or "SPEAKER_1". Match on first name when looking up history; the CLI's --participant flag does fuzzy matching for you.
  • Briefs are sensitive. Always chmod 600. They contain relationship intelligence the user wouldn't want leaked.
  • Be honest about staleness. If the most recent meeting with this person is 3+ months old, lead with that fact: "Last conversation (4 months ago): …" — the user's mental model of the relationship may be more recent than reality.
  • Run the three live-source CLI calls in parallel. They're independent. Synthesize only from returned source-backed results after all three exit successfully.
  • A failed commitment projection is unavailable, not empty. Never claim “No open commitments” from a nonzero graph command.

© silverstein, MIT. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

Just SKILL.md in .opencode/skills/minutes-brief of silverstein/minutes.

Open the folder on GitHubat commit d3285c7

Compare with similar skills

Minutes Brief next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.

Minutes Brief compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Minutes Brief this skillsilverstein/minutes1.5k—~3.1kAutomated safety check: PassMIT
Meeting Minutesgithub/awesome-copilot40k2 repos~2kAutomated safety check: PassMIT
Meeting Minutesaipoch/medical-research-skills1.9k—~1.8kAutomated safety check: PassMIT
Meetingsalirezarezvani/claude-skills28k—~1.5kAutomated safety check: PassMIT
Meeting Minutes Takerdaymade/claude-code-skills1.4k—~9.8kAutomated safety check: PassMIT
Briefalirezarezvani/claude-skills28k—~967Automated safety check: PassMIT

Similar skills

  • Meeting Minutes

    github/awesome-copilot

    Official

    Generate concise, actionable meeting minutes for internal meetings.

    40k GitHub starsUsed in 2 repos~2k tokens
    Productivity & AutomationAuto-check passed
  • Meeting Minutes

    aipoch/medical-research-skills

    Use meeting minutes for other workflows that need structured execution, explicit assumptions, and clear output boundaries.

    1.9k GitHub stars~1.8k tokensUpdated 23 days ago
    Productivity & AutomationAuto-check passed
  • Meetings

    alirezarezvani/claude-skills

    A skill your agent uses when someone wants to decide whether a meeting is worth calling, price a meeting in dollars, build a timeboxed agenda with desired outcomes, or turn messy meeting notes into…

    28k GitHub stars~1.5k tokensUpdated 1 mo ago
    Productivity & AutomationAuto-check passed
  • Meeting Minutes Taker

    daymade/claude-code-skills

    Transforms raw meeting transcripts into structured, evidence-based meeting minutes — merges multiple versions without losing content, reviews existing minutes against the transcript for missing…

    1.4k GitHub stars~9.8k tokensUpdated today
    Productivity & AutomationAuto-check passed
  • Brief

    alirezarezvani/claude-skills

    /cs:brief <topic — Generate a one-page strategy brief from an office-hours intake.

    28k GitHub stars~967 tokensUpdated 1 mo ago
    Marketing & SEOAuto-check passed
  • Meeting Minutes Generator

    aipoch/medical-research-skills

    Generates structured meeting minutes from text transcripts. An agent skill from aipoch/medical-research-skills.

    1.9k GitHub stars~1.5k tokensUpdated 23 days ago
    Productivity & AutomationAuto-check passed

More from silverstein/minutes

All 24 skills in this repo
  • Minutes Video Review

    silverstein/minutes

    Analyze a product walkthrough, bug report video, Loom, or ScreenPal using Minutes transcription plus visual review.

    1.5k GitHub stars~1.4k tokensUpdated yesterday
    Auto-check: notes
  • Minutes Mirror

    silverstein/minutes

    Self-coaching analysis of your own behavior across meetings — talk-time ratio, filler words, hedging language, monologue length, energy patterns, and (when meetings are tagged via /minutes-tag) what…

    1.5k GitHub stars~3.5k tokensUpdated yesterday
    Auto-check passed
  • Minutes Tag

    silverstein/minutes

    Lightweight outcome tagging for meetings — won, lost, stalled, great, or noise.

    1.5k GitHub stars~2.2k tokensUpdated yesterday
    Auto-check passed
  • Minutes Verify

    silverstein/minutes

    Verify that Minutes is properly set up and working — model downloaded, mic accessible, directories exist, no stale state.

    1.5k GitHub stars~801 tokensUpdated yesterday
    Auto-check passed
  • Minutes Record

    silverstein/minutes

    Start or stop recording a meeting, call, or voice memo. An agent skill from silverstein/minutes.

    1.5k GitHub stars~1.1k tokensUpdated yesterday
    Auto-check passed
  • Minutes Cleanup

    silverstein/minutes

    Manage old recordings — find large files, archive old meetings, delete processed originals.

    1.5k GitHub stars~852 tokensUpdated yesterday
    Auto-check passed

Questions about Minutes Brief

What does Minutes Brief do?

Fast non-interactive briefing before any meeting — auto-detects your next calendar event, pulls relationship history, surfaces open commitments, and produces a one-page brief in under 30 seconds. Minutes Brief is an agent skill from silverstein/minutes. Fast non-interactive briefing before any meeting — auto-detects your next calendar event, pulls relationship history, surfaces open commitments, and produces a one-page brief in under 30 seconds.

When should I use Minutes Brief?

Minutes Brief fits situations like: give me a quick brief; whats coming up; background on my next call; who am I meeting next.

How do I install Minutes Brief in Claude Code?

Run `npx skills add silverstein/minutes --skill minutes-brief -a claude-code`. Or copy the skill folder (.opencode/skills/minutes-brief in silverstein/minutes) into .claude/skills/minutes-brief in your project. Claude Code loads it when a task matches its description.

How do I install Minutes Brief in Codex?

Run `npx skills add silverstein/minutes --skill minutes-brief -a codex`. Or copy the skill folder (.opencode/skills/minutes-brief in silverstein/minutes) into .agents/skills/minutes-brief in your project. Codex loads it when a task matches its description.

Can I use Minutes Brief in Cursor, Gemini CLI or GitHub Copilot?

Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add silverstein/minutes --skill minutes-brief -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/minutes-brief, .gemini/skills/minutes-brief, .github/skills/minutes-brief and .opencode/skills/minutes-brief in your project.

What does Minutes Brief need to run?

Going by SKILL.md and its folder, Minutes Brief needs the command-line tools its instructions call (node, git and osascript). Compatibility (from SKILL.md): opencode.

Does Minutes Brief access the network?

SKILL.md contains no URLs. Its commands use git, which can reach the network depending on how they are called. This is read from the text; nothing was executed.

Is Minutes Brief safe to install?

Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.

What licence does Minutes Brief use?

Minutes Brief is published under the MIT licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does Minutes Brief use?

About 3.1k tokens (SKILL.md is roughly 12k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full.

What are the alternatives to Minutes Brief?

Skills that share tags, products or a category with Minutes Brief: Meeting Minutes (github/awesome-copilot, 40k stars), Meeting Minutes (aipoch/medical-research-skills, 1.9k stars), Meetings (alirezarezvani/claude-skills, 28k stars) and Meeting Minutes Taker (daymade/claude-code-skills, 1.4k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Minutes Brief?

silverstein (a GitHub user) maintains it in silverstein/minutes, which has 1,542 GitHub stars. The repository holds 24 skills in this directory. The repository was last updated on October 8, 2026.

Source: silverstein/minutes on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.