7 newsletters since yesterday, one line each
7 newsletters since yesterday at 12:30. The Pricing Letter is the one worth reading in full today.
Weekdays at lunch: one line per newsletter that arrived since yesterday, with the single takeaway worth keeping and which issue to read in full.
Reads
You get
The card on the right is a sample. Yours shows your own numbers and updates in place every run.
What lands on your board
7 newsletters since yesterday at 12:30. The Pricing Letter is the one worth reading in full today.
Subscribing is easy and reading is not. Most founders follow a dozen newsletters on pricing, fundraising, growth and their own market, and most issues sit unread under a growing count until a weekend purge.
The agent reads every issue that came in since yesterday and writes one line for each: what it says that you would want to remember. Three newsletters on the same story become one line. It also names the issue worth opening in full, which on most days is none.
Search users.messages.list with q: "newer_than:1d -from:me" (or your own label:newsletters) and keep messages that carry a List-Id or List-Unsubscribe header, read with format: metadata.
Drop receipts, shipping updates, password resets and pure sales promotions (a discount code and a product grid). Keep issues with an author and an argument.
Fetch with format: full, decode the text/plain part from base64url (or strip the text/html part), and cut the footer, sponsor blocks and "forwarded by a friend" boilerplate.
For every issue, the sender and the single takeaway in under 25 words, specific enough to act on. Merge issues covering the same story into one line that names each source.
If one issue deserves a full read, say which and why in a sentence. Mark issues as read only if asked.
With BRIEFKIN_TOKEN set, the agent sends the result as one card that updates in place every run. Without it, you get the report in the chat.
Pick your agent, where the data comes from and how to schedule it; the commands below follow.
One command. It puts the SKILL.md in ~/.claude/skills and in ~/.agents/skills, which Codex, Cursor, Gemini CLI, Copilot CLI and OpenCode read too.
shellcurl -fsSL https://briefkin.dev/skills/gmail-newsletter-digest/install | shAsk Claude Code:
promptRun the gmail-newsletter-digest skillOr paste the prompt instead of installing:
promptLook through my Gmail for newsletters that arrived since yesterday at this time: messages with a List-Id or List-Unsubscribe header, but not receipts, shipping notices or pure sales promotions. Read each issue and give me one line per newsletter: the sender and the single takeaway worth remembering, in under 25 words and specific enough to act on. If several newsletters cover the same story, merge them into one line that names each. End by telling me which single issue, if any, is worth reading in full and why. Don't mark anything read or archive anything.Weekdays at 12:30, from cron on any machine where Claude Code is signed in:
crontab30 12 * * 1-5 claude -p "Run the gmail-newsletter-digest skill"Add it with crontab -e. A laptop asleep at that time runs it when it wakes if you use launchd or Task Scheduler instead; GitHub Actions works for the API route.
Connect Briefkin once with curl -fsSL https://briefkin.dev/install | sh and every run lands on your board as a report card. How setup works
SKILL.md---
name: gmail-newsletter-digest
description: "Weekdays at lunch: one line per newsletter that arrived since yesterday, with the single takeaway worth keeping and which issue to read in full. Use when the user asks for this, or when it runs on a schedule."
---
# Gmail newsletter digest
Weekdays at lunch: one line per newsletter that arrived since yesterday, with the single takeaway worth keeping and which issue to read in full.
Subscribing is easy and reading is not. Most founders follow a dozen newsletters on pricing, fundraising, growth and their own market, and most issues sit unread under a growing count until a weekend purge.
The agent reads every issue that came in since yesterday and writes one line for each: what it says that you would want to remember. Three newsletters on the same story become one line. It also names the issue worth opening in full, which on most days is none.
## Two ways to get the data
Use the API or MCP route when you have an MCP server, a CLI or an API token for the app. Otherwise, if you can drive a browser the user is signed in to (Claude in Chrome, a Playwright or Chrome DevTools MCP server), use the browser route. Say which one you used.
## Route A: API or MCP server
### What you need
- Read access to Gmail through a connector or MCP server, or the Gmail API with the `gmail.readonly` scope.
If something is missing, say what and stop; do not invent numbers.
### Steps
1. **Find the newsletters** (Gmail): Search `users.messages.list` with `q: "newer_than:1d -from:me"` (or your own `label:newsletters`) and keep messages that carry a `List-Id` or `List-Unsubscribe` header, read with `format: metadata`.
2. **Leave out what isn't editorial**: Drop receipts, shipping updates, password resets and pure sales promotions (a discount code and a product grid). Keep issues with an author and an argument.
3. **Read each issue** (Gmail): Fetch with `format: full`, decode the `text/plain` part from base64url (or strip the `text/html` part), and cut the footer, sponsor blocks and "forwarded by a friend" boilerplate.
4. **Write one line each**: For every issue, the sender and the single takeaway in under 25 words, specific enough to act on. Merge issues covering the same story into one line that names each source.
5. **Pick the one to read**: If one issue deserves a full read, say which and why in a sentence. Mark issues as read only if asked.
## Route B: the browser
### What you need
- Signed in to Gmail in the browser, with the inbox in its usual tabbed layout or with your newsletter label set up.
If something is missing, say what and stop; do not invent numbers.
### Steps
Start at https://mail.google.com/. The user is signed in already; never type passwords or codes, and ask if a sign-in page appears.
1. **Find the issues**: Search `newer_than:1d -from:me` (or `label:newsletters newer_than:1d` if you keep a label) and keep the messages that show an **Unsubscribe** link next to the sender name; that link is Gmail's sign of a `List-Unsubscribe` header. The **Promotions** and **Updates** tabs hold most of them.
2. **Leave out the non-editorial mail**: Skip receipts, shipping notices, password resets and sales mail that is a product grid with a discount code. Keep issues with an author and an argument.
3. **Read each issue**: Open the message, click **Show trimmed content** if the issue is cut off, and read it ignoring the sponsor blocks and the footer. Do not click links in the mail.
4. **Write one line each and pick one**: For every issue, the sender and the single takeaway in under 25 words. Merge issues that cover the same story into one line naming each source. Say which issue, if any, is worth reading in full and why. Opening an issue marks it read; use **Mark as unread** on it afterwards if you want the inbox untouched.
Note: Opening a message in the browser marks it read, unlike the API route, which reads without changing state. Gmail's **Manage subscriptions** page (in the left menu) lists senders by frequency and is a quicker place to see what you never open.
## The report
- One line per newsletter since yesterday: the sender and the takeaway.
- Stories covered by several newsletters, merged into one line.
- The one issue worth reading in full, if there is one.
## Post it to Briefkin
When `BRIEFKIN_TOKEN` is set, or `~/.config/briefkin/env` exists (the installer from https://briefkin.dev/docs writes it), also send the result to the user's Briefkin board as one card. Keep the `item_key` the same on every run so the card updates in place, use a new `event_id` per run, and put today's real numbers in place of the example ones. If the request fails, say so once and carry on.
```bash
[ -n "${BRIEFKIN_TOKEN:-}" ] || . ~/.config/briefkin/env
curl -sS -X POST "${BRIEFKIN_URL:-https://briefkin.dev}/v1/events" \
-H "authorization: Bearer $BRIEFKIN_TOKEN" \
-H "content-type: application/json" \
-d @- <<'JSON'
{
"v": 1,
"event_id": "gmail-newsletter-digest/<date of this run, e.g. 2026-10-16>",
"item_key": "gmail-newsletter-digest",
"project": "Inbox",
"source": "Gmail newsletter digest skill",
"kind": "update",
"title": "7 newsletters since yesterday, one line each",
"summary": "**7** newsletters since yesterday at 12:30. The Pricing Letter is the one worth reading in full today.",
"widget": {
"type": "KeyValue",
"props": {
"items": [
{
"key": "The Pricing Letter",
"value": "Usage plans with a minimum commit renewed 9 points better than pure pay-as-you-go across 300 SaaS companies."
},
{
"key": "SaaS Weekly",
"value": "Median seed rounds closed in 11 weeks this quarter, down from 14; most included a SAFE extension."
},
{
"key": "Growth Notes",
"value": "Moving the signup form above the fold on pricing pages lifted trial starts 12% in two A/B tests."
},
{
"key": "Infra Digest",
"value": "A teardown of halving a $9k monthly database bill with read replicas and connection pooling."
},
{
"key": "Design Signals",
"value": "Empty states that show sample data cut first-week drop-off in three onboarding teardowns."
},
{
"key": "Founder Field Notes",
"value": "Write the board update before the metrics review: it forces you to name the one decision you need."
},
{
"key": "Market Map",
"value": "Two scheduling startups raised Series A rounds this month, both pitching an AI assistant as the main feature."
}
]
}
}
}
JSON
```
Widgets come from the Briefkin catalog (https://briefkin.dev/v1/catalog). Without a token, just give the user the report.
## Run it on a schedule
Suggested: Weekdays at 12:30 (`30 12 * * 1-5`). From cron, with the agent signed in on that machine:
```bash
30 12 * * 1-5 claude -p "Run the gmail-newsletter-digest skill"
# Codex: codex exec, Gemini CLI: gemini -p, Copilot CLI: copilot -p, OpenCode: opencode run
```
Or use the agent's own scheduler (Claude Code routines, Codex automations, Claude or ChatGPT scheduled tasks): https://briefkin.dev/skills/gmail-newsletter-digest#schedule
Source: https://briefkin.dev/skills/gmail-newsletter-digest
Read access to Gmail through a connector or MCP server, or the Gmail API with the gmail.readonly scope.
https://gmailmcp.googleapis.com/mcp/v1, in developer preview with the other Google Workspace MCP servers. It needs an OAuth client in Google Cloud set up for an AI agent (setup).gmail.readonly scope, and store GMAIL_CLIENT_ID, GMAIL_CLIENT_SECRET and GMAIL_REFRESH_TOKEN. Publish the OAuth app (or make it Internal in Workspace): refresh tokens from an app left in Testing expire after seven days.gmail.readonly can search and read but not send, delete or relabel; add gmail.compose only if you want the agent to leave drafts.
Both carry a List-Unsubscribe header, so that alone isn't enough. The agent keeps issues with editorial text, an author and an argument, and drops mail that is mostly a product grid, a discount code or a transactional notice. If it gets a sender wrong, name it in the prompt as always keep or always skip.
It can list the newsletters you have skipped for two weeks with the unsubscribe link from their List-Unsubscribe header, but it doesn't click them. Unsubscribing stays your call.
No. The skill runs with your own agent and gives you the report in the chat. With a Briefkin token set, each run also posts it to your board as a report card, so you can glance at it instead of asking again.
Claude Code, Codex, Cursor, Gemini CLI, GitHub Copilot CLI, OpenCode, Windsurf and Amp all read SKILL.md files. The install command puts it in ~/.claude/skills and ~/.agents/skills, where they look. ChatGPT, the Claude app and Grok take the prompt instead. Pick your agent under "Install and run it" and the commands change to match.
Yes. Switch the data source to "Your browser" and the steps change to the Gmail pages themselves: the agent opens mail.google.com in a browser you are signed in to and reads the numbers off the page. Claude Code, Codex, Cursor, Claude app and ChatGPT can drive a browser out of the box; Gemini CLI, Copilot CLI and OpenCode need a Playwright or Chrome DevTools MCP server first. It is slower than the API and needs you signed in, so it suits a report you run by hand or from a scheduler on your own computer.
Skills are free to use with any agent. Briefkin is where their results land: one board for every report, number and question your agents send, with a ping when something needs you.
Free during the open beta. No card.