Developer Community Listening — Stack Exchange, Lemmy, Mastodon

Find out where developers are talking about your product — and go read it. One search term, three communities, one flat schema: every mention arrives with its text, a direct link, the author, the community it came from and that platform's engagement numbers.

• 🎧 Three sources, one run — Stack Exchange (any site in the network), Lemmy, and Mastodon hashtags
• 🔗 Every row is actionable — direct URL and author handle on every mention, so you can reply, not just count
• 📊 Engagement attached — votes, answers and views on Stack Exchange; score and comments on Lemmy; favourites, boosts and follower counts on Mastodon
• 🆓 No key required — official and open APIs; an optional free Stack Exchange key just raises the daily quota

Quick start (API)

Everything said about "supabase" in the last 30 days, across all three:

curl -X POST "https://api.apify.com/v2/acts/foxlabs~community-listening/run-sync-get-dataset-items?token=YOUR_APIFY_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{ "query": "supabase", "datePreset": "last_30_days", "maxResults": 300 }'

Prefer no code? Open the Input tab, type your brand, press Start.

What you get

Field	Type	Description
source	string	stackexchange, lemmy or mastodon
community	string	The site, Lemmy community or hashtag+instance the mention came from
title	string	Post/question title (Mastodon posts have none)
text	string	The body, HTML stripped, capped at 4,000 characters
url	string	Direct link to the mention
linkUrl	string	The external link a Lemmy post points at, when it has one
author / authorUrl	string	Who wrote it and their profile
authorReputation / authorFollowers	number	Stack Exchange reputation / Mastodon follower count
score	number	Votes (Stack Exchange), score (Lemmy), favourites (Mastodon)
comments	number	Answers (Stack Exchange), comments (Lemmy), replies (Mastodon)
views	number	Stack Exchange view count
reblogs	number	Mastodon boosts
isAnswered	bool	Whether the Stack Exchange question has an accepted answer
tags	array	Question tags / post hashtags
createdAt / lastActivityAt	string	When it was posted / last touched (ISO)
matchedQuery	string	The term this run was listening for
license	string	Attribution string on Stack Exchange rows (CC BY-SA 4.0)

The run also writes a per-source report to the key-value store (SOURCE_REPORT) with the count from each source — so you always know whether "few results" means a quiet week or a source that failed.

Sample output

{
  "source": "stackexchange",
  "community": "stackoverflow",
  "title": "Supabase RLS policy blocks insert from edge function",
  "text": "I have a policy that should allow inserts for authenticated users…",
  "url": "https://stackoverflow.com/questions/79000000/supabase-rls-policy-blocks-insert",
  "author": "Arnav Sharma",
  "authorReputation": 1,
  "score": 2,
  "comments": 1,
  "views": 61,
  "isAnswered": false,
  "tags": ["next.js", "supabase"],
  "createdAt": "2026-07-31T10:22:41.000Z",
  "matchedQuery": "supabase",
  "license": "Stack Exchange content, CC BY-SA 4.0"
}

Download as JSON, CSV, Excel or HTML, or read it from the API.

Options

• Search term — the brand, product or topic. Required.
• Sources — any combination of Stack Exchange, Lemmy, Mastodon.
• Time window — 7 / 30 / 90 days, a year, or all time. All time is the default, combined with newest-first sorting: Stack Overflow's volume for a single product name can be as low as a couple of questions a month (measured: 2 for "supabase" in 30 days), so a narrow window can return almost nothing even though the term is discussed. You still get the newest mentions first.
• Sort by — newest (monitoring), most votes (what mattered), recent activity (still moving).
• Stack Exchange sites — search several sites in one run (stackoverflow, serverfault, dba, security, …).
• Lemmy / Mastodon instance — point at a niche server if your audience lives there.
• Mastodon hashtags — override the tag(s); defaults to the search term.
• Stack Exchange key — optional, raises the daily quota from ~300 to 10,000 requests.

Example inputs (copy & paste)

// 1) Weekly brand monitor
{ "query": "clickhouse", "datePreset": "last_7_days" }

// 2) Where the pain is — unanswered questions about your product
{ "query": "supabase", "sources": ["stackexchange"], "sortBy": "newest", "maxResults": 1000 }

// 3) Ops-side chatter, not just app developers
{ "query": "kubernetes operator", "sources": ["stackexchange"], "stackExchangeSites": ["serverfault", "devops", "stackoverflow"] }

// 4) The fediverse view of a topic
{ "query": "selfhosted", "sources": ["lemmy", "mastodon"], "mastodonHashtags": ["selfhosted", "selfhosting"] }

// 5) Competitor watch, most-discussed first
{ "query": "planetscale", "sortBy": "votes", "datePreset": "last_90_days" }

Use cases

• Developer marketing / DevRel. See every question, post and toot about your product this week, with a link to reply to. Unanswered Stack Exchange questions are the highest-value ones — they're a customer stuck in public.
• Competitor monitoring. Run the same query against a competitor's name and compare volume, sentimentally-loaded titles and which communities they show up in.
• Product & docs feedback. Recurring question tags tell you which part of your product the docs fail to explain.
• Community discovery. Find the Lemmy communities and Mastodon hashtags where your audience actually gathers, before you try to participate.
• Weekly digests. Schedule it, export to a sheet or Slack, and read the week's mentions in one place.

Performance & cost

Plain HTTP against official/open APIs — no browser, no proxy, no third-party fees. Stack Exchange returns 100 results per page with deep paging; Lemmy pages 20 at a time; Mastodon hashtag timelines page by max_id. Quotas, honestly: Stack Exchange allows ~300 requests/day per IP without a key (a free key raises it to 10,000/day), and the Actor honours the API's own backoff requests. Lemmy and Mastodon are queried at a polite pace.

Integrations

JavaScript (apify-client):

import { ApifyClient } from 'apify-client';
const client = new ApifyClient({ token: 'YOUR_APIFY_TOKEN' });
const run = await client.actor('foxlabs/community-listening').call({
  query: 'clickhouse', datePreset: 'last_7_days',
});
const { items } = await client.dataset(run.defaultDatasetId).listItems();

Python (apify-client):

from apify_client import ApifyClient
client = ApifyClient("YOUR_APIFY_TOKEN")
run = client.actor("foxlabs/community-listening").call(run_input={
    "query": "supabase", "sources": ["stackexchange"], "sortBy": "newest",
})
for item in client.dataset(run["defaultDatasetId"]).iterate_items():
    print(item["source"], item.get("score"), item["url"])

Also works with Make / n8n / Zapier, scheduled runs, webhooks and the Apify MCP server.

Honest limits (please read)

• Mastodon is hashtag-based here. Its full-text search requires an account token, so this Actor reads hashtag timelines instead. A post that mentions your product without a hashtag will not appear. Set mastodonHashtags to the tags your community actually uses.
• Lemmy is per-instance. The Fediverse has no global index; searching lemmy.world sees that instance's view. Point at another instance to cover a different corner.
• Stack Exchange covers questions, not comments — and its volume per product is thinner than people expect. Measured on 2026-08-01: "supabase" returned 100 matching questions all-time but only 2 in the last 30 days. That's the platform, not the Actor; keep the window wide and sort newest-first for monitoring.
• Reddit is deliberately not a source. Its robots.txt disallows crawling and its public-content policy requires a contract for commercial use, so including it would mean breaking those terms. This Actor stays on sources whose terms permit it.
• Full-text search matches URLs too. A product name that is also a hosting domain (supabase.co, vercel.app) will occasionally match a post that merely links to an asset there. Measured example in Data quality below.
• If a source fails mid-run, the others still return and SOURCE_REPORT records the failure — you never get a silently short list.

Data quality (measured 2026-08-01)

A 300-mention run for "supabase" across all three sources, 100 from each:

Check	Result
url, text, author, community, createdAt, score, comments	100% on all three sources
Unique mentions (by URL)	300 / 300
Term actually present in the row	300 / 300
Links verified against the source API (title + link match)	10 / 10
Stack Exchange attribution string	100 / 100 rows
Columns that were empty for every row	none

One honest caveat about matching: these are full-text searches, so a term can match inside a URL. One Lemmy post in the measured run was a biography of Mary "Mother" Jones — it matched because its image is hosted on …supabase.co. Technically a match, not a mention. Scan the text field before acting on results for brand names that double as hosting domains.

Pricing

$0.002 per mention ($2 per 1,000) — pay only for results. You are charged per mention actually returned. A run that finds nothing costs nothing beyond a few seconds of compute.

Pay per result — billed per mention returned, with an Apify free tier to evaluate. The APIs behind it are free, so there are no proxy or third-party costs.

FAQ

Which sources does it cover?

The Stack Exchange network (Stack Overflow, Server Fault, DBA, Security and 100+ others), Lemmy, and Mastodon hashtag timelines.

Do I need any API key?

No. A free Stack Exchange key is optional and only raises the daily quota.

Why isn't Reddit included?

Because Reddit's own rules don't allow it — see Honest limits. We'd rather ship three sources you can use than a fourth that puts your account at risk.

Can I track several brands?

Run it once per term (that keeps matchedQuery clean), or schedule several runs and merge the datasets.

How do I find unanswered questions?

Use Stack Exchange as the only source and filter the output on isAnswered = false — those are the customers stuck in public.

Is the data licensed for reuse?

Stack Exchange content is CC BY-SA 4.0 and each row carries the attribution string. Lemmy and Mastodon posts belong to their authors; treat them as you would any public post.

How fresh is it?

Live — all three APIs return current data at request time. Schedule the Actor to build a history.

Troubleshooting

• 0 mentions → widen the time window, try a broader term, or check SOURCE_REPORT to see whether a source returned nothing or failed.
• Nothing from Mastodon → your term may not be used as a hashtag; set mastodonHashtags explicitly.
• Stack Exchange stops early → you've hit the ~300/day keyless quota. Add a free key in Advanced.
• Nothing from Lemmy → that instance may not host communities discussing your topic; try another instance.

Support

Questions, another source you'd like added, or a custom build? Open the Issues tab, or e-mail info@foxlabs.com.tr. We reply fast.

If this Actor saves you time, a ⭐ review really helps.

Changelog

0.1 — 2026-08-01

• Initial release: Stack Exchange + Lemmy + Mastodon in one schema, with per-source reporting, date windows, multi-site Stack Exchange search and configurable Lemmy/Mastodon instances.

---

Part of the foXLabs data platform — developer, company and market intelligence scrapers. See also Hacker News Search & Who-is-Hiring and GitHub Trending & Rising Repos for the rest of the developer-community picture.