What it does
The week after a launch, a product manager checks the same chart several times a day: is anybody using it? The honest version of that number is a share, not a count. Two hundred users of the new export means one thing on a quiet Sunday and something else on a Tuesday with twice the traffic.
So the agent divides the people who used the feature by the people active that day, and posts the result every morning with one point per day since launch. If the feature sits behind a flag, it counts only users who had the flag on, so a 20% rollout does not read as 80% indifference.
How it works
- 1
1Name the feature
The event that means the feature was used (for example
report exportedwithproperties.format = 'pdf'), the launch date, and the feature flag key if there is one. 2Count daily feature users
HogQL through
POST /api/projects/:project_id/query/:SELECT toDate(timestamp) AS day, count(DISTINCT person_id) FROM events WHERE event = 'report exported' AND timestamp >= '2026-10-07' GROUP BY day ORDER BY day.3Count daily active users
The same query over your active events (
$pageview,$screenor your key event). For a flagged feature, restrict both queries to events where the$feature/<flag-key>property is true.- 4
4Work out adoption
Daily adoption is feature users divided by active users for each day. Also compute reach since launch (distinct people who used it over distinct people active since launch) and repeat use (feature users seen on two or more days).
- 5
5Post yesterday's share
Headline: yesterday's adoption in percent, one history point per day since launch. Add reach, repeat use and the day-by-day table.
Post it to the board
With
BRIEFKIN_TOKENset, the agent sends the result as one card that updates in place every run. Without it, you get the report in the chat.
Install and run it
Pick your agent, where the data comes from and how to schedule it; the commands below follow.
Install the skill
One command. It puts the SKILL.md in ~/.claude/skills and in ~/.agents/skills, which Codex, Cursor, Gemini CLI, Copilot CLI and OpenCode read too.
shell
curl -fsSL https://briefkin.dev/skills/posthog-feature-adoption/install | shRun it
Ask Claude Code:
prompt
Run the posthog-feature-adoption skillOr paste the prompt instead of installing:
prompt
Using PostHog HogQL queries, for each day since we launched [feature] on [date]: count distinct people who fired [feature event] and distinct people who were active ($pageview or $screen), restricted to users with the feature flag [flag key] on if it is flagged. Give me daily adoption (feature users divided by active users) as a table by day since launch, yesterday's figure, cumulative reach since launch, and the share of feature users who used it on two or more days. Tell me whether adoption is still rising or has flattened.Put it on a schedule
Daily at 9:00, from cron on any machine where Claude Code is signed in:
crontab
0 9 * * * claude -p "Run the posthog-feature-adoption 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.Send it to your board
Connect Briefkin once with
curl -fsSL https://briefkin.dev/install | shand every run lands on your board as a stat tile. How setup works
View the SKILL.mdRaw file
SKILL.md---
name: posthog-feature-adoption
description: "Every morning after a launch: the share of active users who used the new feature, by day since launch, from PostHog. Use when the user asks for this, or when it runs on a schedule."
---
# PostHog feature adoption
Every morning after a launch: the share of active users who used the new feature, by day since launch, from PostHog.
The week after a launch, a product manager checks the same chart several times a day: is anybody using it? The honest version of that number is a share, not a count. Two hundred users of the new export means one thing on a quiet Sunday and something else on a Tuesday with twice the traffic.
So the agent divides the people who used the feature by the people active that day, and posts the result every morning with one point per day since launch. If the feature sits behind a flag, it counts only users who had the flag on, so a 20% rollout does not read as 80% indifference.
## 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
- A PostHog personal API key with `query:read`, the project ID and host (`POSTHOG_API_KEY`, `POSTHOG_PROJECT_ID`, `POSTHOG_HOST`), or a PostHog connector.
- The feature's event name, its launch date and its feature flag key, if any.
If something is missing, say what and stop; do not invent numbers.
### Steps
1. **Name the feature**: The event that means the feature was used (for example `report exported` with `properties.format = 'pdf'`), the launch date, and the feature flag key if there is one.
2. **Count daily feature users** (PostHog): HogQL through `POST /api/projects/:project_id/query/`: `SELECT toDate(timestamp) AS day, count(DISTINCT person_id) FROM events WHERE event = 'report exported' AND timestamp >= '2026-10-07' GROUP BY day ORDER BY day`.
3. **Count daily active users** (PostHog): The same query over your active events (`$pageview`, `$screen` or your key event). For a flagged feature, restrict both queries to events where the `$feature/<flag-key>` property is true.
4. **Work out adoption**: Daily adoption is feature users divided by active users for each day. Also compute reach since launch (distinct people who used it over distinct people active since launch) and repeat use (feature users seen on two or more days).
5. **Post yesterday's share**: Headline: yesterday's adoption in percent, one history point per day since launch. Add reach, repeat use and the day-by-day table.
## Route B: the browser
### What you need
- Signed in to PostHog as a member of the project; SQL insights are available on every plan.
If something is missing, say what and stop; do not invent numbers.
### Steps
Start at https://us.posthog.com/. The user is signed in already; never type passwords or codes, and ask if a sign-in page appears.
1. **Open a SQL insight**: PostHog URLs include the project ID, so start at the home page (`eu.posthog.com` for EU projects), pick the project, open **Product analytics**, click **New** and choose the **SQL** tab.
2. **Count daily feature users**: Run `SELECT toDate(timestamp) AS day, count(DISTINCT person_id) FROM events WHERE event = 'report exported' AND timestamp >= '2026-10-07' GROUP BY day ORDER BY day`, with the launch date and event name swapped for yours. For a flagged feature add `AND properties.$feature/<flag-key> = 'true'` with the flag key in backticks.
3. **Count daily active users**: Change the `event =` clause to `event IN ('$pageview', '$screen')`, keep the same date and flag conditions, and run again. Copy both result tables.
4. **Work out adoption**: Divide feature users by active users for each day. Run two more queries since launch: distinct people who used the feature over distinct people active (reach), and the share of feature users seen on two or more days (repeat use).
5. **Write it up**: Yesterday's adoption, the day-by-day table, reach and repeat use, and whether the line is still climbing or has flattened.
Note: Each query runs on its own, so a morning check is four queries. The SQL editor keeps the last query in the tab only until you leave, so save the insight to avoid retyping it.
## The report
- **Adoption yesterday**: the share of active users who used the feature.
- **Day by day since launch**, with active users and feature users.
- **Reach since launch** and **repeat use**.
- One line on whether adoption is still climbing or has flattened.
## Post it to Briefkin
When `BRIEFKIN_TOKEN` is set, or `~/.config/briefkin/env` exists (the installer from https://briefkin.dev/docs writes it), also send the result to the user's Briefkin board as one card. Keep the `item_key` the same on every run so the card updates in place, use a new `event_id` per run, and put today's real numbers in place of the example ones. If the request fails, say so once and carry on.
```bash
[ -n "${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": "posthog-feature-adoption/<date of this run, e.g. 2026-10-16>",
"item_key": "posthog-feature-adoption",
"project": "acme-web",
"source": "PostHog feature adoption skill",
"kind": "metric",
"title": "PDF export used by 14.2% of active users yesterday",
"metric": {
"label": "Daily adoption, PDF export",
"value": 14.2,
"unit": "%",
"better": "higher",
"period_start": "<start of the period, ISO 8601>",
"period_end": "<end of the period, ISO 8601>"
},
"glance": {
"type": "Grid",
"props": {
"columns": 3
},
"children": [
{
"type": "Stat",
"props": {
"label": "Used it yesterday",
"value": 1184
}
},
{
"type": "Stat",
"props": {
"label": "Reach since launch",
"value": 27,
"unit": "%"
}
},
{
"type": "Stat",
"props": {
"label": "Used on 2+ days",
"value": 41,
"unit": "%"
}
}
]
},
"widget": [
{
"type": "Table",
"props": {
"columns": [
"Day",
"Active users",
"Used PDF export",
"Adoption"
],
"align": [
"left",
"right",
"right",
"right"
],
"rows": [
[
"1, Wed Oct 7",
8020,
249,
"3.1%"
],
[
"2, Thu Oct 8",
8210,
558,
"6.8%"
],
[
"3, Fri Oct 9",
8105,
762,
"9.4%"
],
[
"4, Sat Oct 10",
4310,
440,
"10.2%"
],
[
"5, Sun Oct 11",
3960,
345,
"8.7%"
],
[
"6, Mon Oct 12",
8290,
987,
"11.9%"
],
[
"7, Tue Oct 13",
8410,
1060,
"12.6%"
],
[
"8, Wed Oct 14",
8375,
1097,
"13.1%"
],
[
"9, Thu Oct 15",
8338,
1184,
"14.2%"
]
]
}
},
{
"type": "Callout",
"props": {
"text": "Adoption has risen on every weekday since launch, and 41% of the people who tried it came back on a second day.",
"tone": "ok"
}
}
],
"size": "md"
}
JSON
```
Widgets come from the Briefkin catalog (https://briefkin.dev/v1/catalog). Without a token, just give the user the report.
Set `period_start` and `period_end` to the window the number covers: the tile names it from them ("Last 7 days", "Today", "This month"). For a count right now (open items, a balance), send `"period_label": ""` so no period shows; for another name, send it as `period_label` (at most 24 characters).
## Run it on a schedule
Suggested: Daily at 9:00 (`0 9 * * *`). From cron, with the agent signed in on that machine:
```bash
0 9 * * * claude -p "Run the posthog-feature-adoption 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/posthog-feature-adoption#schedule
Source: https://briefkin.dev/skills/posthog-feature-adoption
What your agent needs
A PostHog personal API key with
query:read, the project ID and host (POSTHOG_API_KEY,POSTHOG_PROJECT_ID,POSTHOG_HOST), or a PostHog connector.The feature's event name, its launch date and its feature flag key, if any.
Connecting PostHog
- PostHog MCP server, PostHog's official remote server at
https://mcp.posthog.com/mcp. Sign in with OAuth, or let the PostHog Wizard install it in your client; it routes to the US or EU cloud based on your account. Clients without OAuth can send a personal API key made with the MCP Server preset as a Bearer token (docs). - Query API with a personal API key. Create one in your PostHog account settings, store it in
POSTHOG_API_KEYwithPOSTHOG_PROJECT_IDandPOSTHOG_HOST(https://us.posthog.comorhttps://eu.posthog.com), and give it thequery:readscope, plusinsight:readorfeature_flag:readif the report uses them.
Add ?readonly=true to the MCP URL (or send x-posthog-read-only: true) to drop every tool that creates, updates or deletes; on the API, read scopes alone do the same.
Questions
The feature is rolled out to 25% of users. Is adoption out of 25% or 100%?
Out of the people who could see it. When you give the flag key, both counts include only events where $feature/<flag-key> was true, so the share is measured among users who had the feature.
When should I stop running it daily?
When the daily share has been flat for a week or two, adoption has settled. Move the skill to a weekly schedule as a health check, or archive the card.
Do I need Briefkin to use the PostHog feature adoption skill?
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 stat tile, so you can glance at it instead of asking again.
Which agents can run it?
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.
Can it work without an API or MCP server, from my browser?
Yes. Switch the data source to "Your browser" and the steps change to the PostHog pages themselves: the agent opens us.posthog.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.