Mastodon Scraper — Posts, Hashtags, Accounts & Trends (No Login)

Export Mastodon posts, profiles, hashtags and trends to JSON, CSV or Excel — no account, no API key, no proxy. Point it at mastodon.social (the default) or any Mastodon instance in the fediverse and pull clean, structured data in one run. Built on each instance's official public REST API, so there is nothing to bypass and nothing to break.

Mastodon is the largest open, decentralized social network — a fast-growing, ad-free source of real public discourse. This scraper turns it into a dataset you can actually use for social listening, brand monitoring, research and AI/RAG pipelines.

What you can scrape

• Hashtag timeline — every recent public post for any #tag, deep-paginated, with an optional media-only filter. Great for topic tracking and social listening.
• Account — any user's full profile plus their posts, with reply/boost filters. Follower counts, bio, join date, post history.
• Trends — what's trending on the instance right now: trending statuses, hashtags, or news links, with usage stats.
• Search — find accounts or hashtags by keyword (up to 40 matches per term, keyless).

Everything lands in one unified dataset with consistent field names and both raw HTML and clean plain-text bodies — ready for a spreadsheet, a database, or an LLM.

Why this actor

Option	The catch	This actor
Other Mastodon scrapers on the Store	Single mode (hashtags or accounts), single-instance assumptions, or monthly subscriptions	Hashtag + account + trends + search, any instance, pay only per result
The raw Mastodon API	You maintain pagination, HTML stripping, rate-limit backoff and retries yourself	One JSON input, clean flat schema, it handles the messy parts
Login-based scrapers	Need credentials / risk your account	100% keyless — only public data, nothing to get banned

• Never crashes on you. Every network call retries transient errors (429 rate-limits, 5xx, timeouts) with backoff and degrades gracefully on hard errors — a bad hashtag or an unreachable instance is skipped, the run still finishes and delivers everything it gathered.
• Zero proxy cost. Official public API, no anti-bot bypass, no residential proxies to pay for.
• Federation-aware. Not limited to one server — scrape mastodon.social, fosstodon.org, hachyderm.io, or any instance.

Common use cases

• Brand & competitor monitoring — track a #brand hashtag or a competitor's account across the fediverse; watch mentions, sentiment inputs and engagement.
• Social listening — follow a topic (#ai, #climate, #elections) with scheduled daily pulls into a dashboard.
• AI / RAG pipelines — feed fresh, open, public discourse into retrieval and fine-tuning without platform lock-in or login walls.
• Trend monitoring — daily snapshots of trending statuses/tags/links for a newsletter or analytics feed.
• Academic & network research — collect public posts and account metadata at scale for computational social science.

Input examples

Recent posts for two hashtags:

{
  "mode": "hashtag",
  "instance": "mastodon.social",
  "searchTerms": ["ai", "opensource"],
  "maxResultsPerTerm": 500
}

A user's profile + original posts (no replies, no boosts):

{
  "mode": "account",
  "instance": "mastodon.social",
  "accounts": ["Gargron"],
  "excludeReplies": true,
  "excludeReblogs": true,
  "maxResultsPerTerm": 200
}

What's trending right now:

{ "mode": "trends", "trendType": "statuses", "maxResultsPerTerm": 40 }

Find accounts by keyword:

{ "mode": "search", "searchType": "accounts", "searchTerms": ["journalism", "python"] }

Output

One row per result, tagged by type. Load it as JSON, CSV, or Excel from the dataset.

Statuses (type: "status"): id, url, uri, created_at, edited_at, language, visibility, text (plain), content_html, spoiler_text, sensitive, is_reblog, in_reply_to_id, replies_count, reblogs_count, favourites_count, author_id, author_acct, author_username, author_display_name, author_url, author_followers, tags[], media_count, media_urls[], card_url

Accounts (type: "account"): id, acct, username, url, display_name, note (plain), note_html, created_at, last_status_at, bot, locked, discoverable, followers_count, following_count, statuses_count, avatar_url, header_url, fields[]

Tags (type: "tag"): name, url, total_uses, total_accounts, history[]

Links (type: "link"): url, title, description, provider_name, author_name, published_at, image, total_uses, total_accounts, history[]

Example status row:

{
  "type": "status",
  "id": "116906215486227555",
  "url": "https://www.npr.org/2026/07/12/nx-s1-5867763/ai-artificial-intelligence",
  "created_at": "2026-07-12T09:00:00.000Z",
  "text": "Getting campaign text messages ahead of the midterms? There could be an AI bot behind it...",
  "author_acct": "npr_national@robot.villas",
  "author_display_name": "NPR Topics: National",
  "author_followers": 5,
  "reblogs_count": 3,
  "tags": ["ai", "politics", "tech"]
}

Pricing

Pay per result — $0.003 per item. You are charged only for rows actually delivered; nothing else. No monthly fee, no proxy surcharge.

• 1,000 items ≈ $3
• 5,000 items ≈ $15

The run stops automatically when your spending limit is reached — no overruns.

How it works (technical notes)

• Uses each instance's official Mastodon REST API — /api/v1/timelines/tag, /api/v1/accounts, /api/v1/trends/*, /api/v2/search — free, public, no key.
• Resilient by design: browser-style User-Agent, exponential-backoff retries on 429/5xx/network errors (honoring Retry-After), and graceful degradation on hard 4xx. One flaky call never sinks the run.
• Verified pagination: max_id cursoring for timelines and account posts (via Link headers), offset paging for trends.
• Status/account text is returned both as raw content_html and stripped plain text, so it drops straight into analysis or an LLM.

Honest scope & limits

• No global timeline mode. Most instances (including mastodon.social) return 422 for the anonymous public/local timeline. Use hashtag mode for a topic feed — it is the keyless way to get a stream of posts by subject.
• Search is accounts + hashtags, single page (≤40 per term). Mastodon disables search pagination and full-text status search without a login. For posts on a topic, use hashtag mode (which paginates deeply).
• Account mode resolves handles known to the chosen instance. For a remote user, set instance to that user's home server for the fullest history.

Data compliance

This actor reads only public Mastodon data through each instance's official, publicly documented REST API. No login credentials are used and no private content is accessed. Public posts and profiles are intentionally public and suitable for research and business use. Please respect each instance's terms of service and the authors' content.