Stack Overflow Scraper — Q&A, Tags & All 170 Sites | from $1.50/1K No Auth

Used by developer-education platforms, AI coding assistants, and enterprise knowledge-base teams.

Scrape questions, answers, and users from Stack Overflow and all 170+ Stack Exchange sites via the official public API — no proxy, no auth required, no rate-limit tricks.

What You Can Scrape

• Questions — titles, tags, scores, answer counts, full body HTML, owner info, dates
• Answers — body HTML, accepted status, vote score, linked question ID
• Users — display name, reputation, creation date, profile link

All 170+ Stack Exchange network sites are supported: stackoverflow, superuser, serverfault, askubuntu, math, physics, datascience, security, unix, softwareengineering, codereview, and more.

Why Stack Exchange Scraper Beats Alternatives

Feature	This actor	Typical alternatives
parse_confidence field	✅ every row	❌ none
All 170+ SE sites	✅ single input	❌ SO-only
Questions + Answers + Users	✅ all 3	❌ questions only
Official API (no proxy)	✅ $0 infra	❌ proxy required
Tag filter + search + ID fetch	✅ all modes	❌ limited
Body text (stripped HTML)	✅ included	❌ HTML only
Pay per result	✅ PPE	❌ rental / flat fee

How to Use

Scrape questions by tag (most common)

Set Data type = questions, Tags = ["python"], Site = stackoverflow. Combine multiple tags for AND filtering: ["python", "pandas"] returns questions that have both tags.

Full-text search

Set Search queries = ["python asyncio tutorial"]. Uses the official /search/advanced endpoint with relevance ranking. Supports quoted phrases.

Fetch answers for specific questions

Set Data type = answers, Question IDs = ["27435284"] to get all answers for a specific question sorted by votes.

Fetch top users

Set Data type = users, leave User IDs blank → returns users sorted by reputation. Set specific User IDs to fetch exact user profiles.

Output Schema

Every row (question, answer, or user) shares the same flat schema:

Field	Type	Description
site	string	SE site slug (e.g. "stackoverflow")
type	string	"question" | "answer" | "user"
question_id	integer	Question ID (null for users)
answer_id	integer	Answer ID (null for questions/users)
user_id	integer	Owner's user ID
title	string	Question title (null for answers/users)
body_html	string	Full body HTML (null if includeBody=false)
body_text	string	HTML-stripped plain text body
tags	array	Tag list (questions only)
score	integer	Vote score (null for users)
answer_count	integer	Number of answers (questions only)
is_answered	boolean	Has accepted answer (questions only)
is_accepted	boolean	This answer is accepted (answers only)
owner_name	string	Display name of owner
owner_id	integer	User ID of owner
owner_reputation	integer	Reputation at scrape time
creation_date	ISO 8601	When the item was created
last_activity_date	ISO 8601	Last activity
link	string	Canonical URL
query	string	Provenance: the search query or tag
parse_confidence	float	0–1 data quality score
warnings	array	Quality warning labels
scraped_at	ISO 8601	Scrape timestamp

API Key (Optional)

Without an API key: 300 requests/day (shared by IP). With a free API key from stackapps.com: 10,000 requests/day.

Each page of 100 results = 1 API request. For most use cases the anonymous quota is sufficient.

What Does parse_confidence Mean?

Every record gets a quality score from 0.0 to 1.0:

• 1.0 — all expected fields present
• < 0.9 — minor missing fields (no title, no owner)
• < 0.7 — major missing fields (no ID, no creation date)

Filter parse_confidence >= 0.9 for clean downstream processing.

Pricing examples

Run	Items	Cost
100 questions (metadata only)	100	~$0.15
1,000 questions with body text	1,000	~$2.00
500 answers for a tag	500	~$0.75
Daily monitoring (100 new questions)	3,000/mo	~$4.50/mo

You only pay for items successfully pushed to the dataset. Body HTML is billed as an additional event only when non-null.

FAQ

Do I need an API key or proxy? No proxy required — the Stack Exchange API is public. Without an API key you get ~300 requests/day (shared by IP). For production runs get a free key at stackapps.com for 10,000 req/day.

Which Stack Exchange sites are supported? All 170+ sites. Set the site field to the slug (e.g. stackoverflow, superuser, datascience, security). Full list at stackexchange.com/sites.

Can I schedule it to monitor new questions? Yes. Use Apify Schedules + fromDate to pull only questions newer than your last run. Combine with a webhook to push new items into Slack, Airtable, or a database.

What if it returns empty results? Check your site slug is correct and your tags filter is valid (tags are lowercase slugs, e.g. python not Python). For answers mode you must provide questionIds or leave it blank to browse recent answers. Errors are logged to the OUTPUT key-value store.

Pricing

Pricing: $1.50 per 1,000 items. Enable includeBody for full Q&A text — additional $0.50/1k for body content.

No API key required for standard usage (300 req/day anonymous). Supply a free Stack Apps key from stackapps.com for 10,000 req/day.

Sites Reference

Full list at stackexchange.com/sites. Common slugs:

stackoverflow · superuser · serverfault · askubuntu · math · physics · datascience · security · unix · softwareengineering · codereview · stats · dba · networkengineering · devops · gaming · electronics

Use with AI agents (MCP)

This actor is callable as a tool by AI agents (Claude Desktop, Cursor, VS Code, n8n, LangGraph, CrewAI, or any MCP-compatible client) via Apify's hosted Model Context Protocol server. An agent uses it to fetch Stack Overflow and Stack Exchange Q&A mid-conversation — e.g. "find the top-voted Python asyncio questions", "get all answers for question 27435284", or "pull the latest security advisories tagged CVE from security.stackexchange.com".

Point your MCP client at this single tool:

{
  "mcpServers": {
    "apify": {
      "command": "npx",
      "args": [
        "mcp-remote",
        "https://mcp.apify.com/?tools=bovi/stack-exchange-scraper",
        "--header",
        "Authorization: Bearer <YOUR_APIFY_TOKEN>"
      ]
    }
  }
}

Minimal call an agent can make (search mode — set maxItems low to keep responses token-lean):

{ "site": "stackoverflow", "dataType": "questions", "searchQueries": ["python asyncio"], "sort": "votes", "maxItems": 10 }

Returns clean, flat rows the agent can reason over directly:

{
  "site": "stackoverflow",
  "type": "question",
  "question_id": 49005651,
  "title": "How does asyncio actually work?",
  "tags": ["python", "python-asyncio", "python-3.x"],
  "score": 892,
  "answer_count": 7,
  "is_answered": true,
  "body_text": "I've been trying to understand asyncio...",
  "owner_name": "J_H",
  "link": "https://stackoverflow.com/questions/49005651",
  "creation_date": "2018-02-26T00:00:00Z",
  "parse_confidence": 1.0,
  "warnings": []
}

Reliability for agents: data comes from the official Stack Exchange public API (no HTML scraping), so rows don't silently break on site redesigns. Every row includes a parse_confidence score (0–1) — filter parse_confidence >= 0.9 for clean downstream processing. The includeBody: false option halves response size when only titles and metadata are needed. No API key required inside the tool — auth is your Apify token in the client config above.

Integrations

Built for developer-education platforms and AI teams building knowledge bases from Stack Overflow and the broader network — the JSON/dataset output drops into the tools you already run, no glue code:

• n8n / Make / Zapier — trigger a run or pipe every new dataset item into 500+ apps (Google Sheets, Airtable, Slack, HubSpot, your database) with no code: n8n, Make, Zapier.
• Webhooks — fire your own endpoint the moment a run finishes, to push results straight into your pipeline (docs).
• MCP server — expose this actor as a tool to Claude, Cursor, or any MCP client so an AI agent can pull this data mid-conversation (guide).
• API & SDKs — fetch the dataset as JSON, CSV, or Excel through the Apify REST API or the Python / JS SDKs.

See all Apify integrations.

Disclaimer

Not affiliated with Stack Exchange Inc. Uses the official public Stack Exchange API under its terms of service. Data is licensed CC BY-SA 4.0 by contributors.

Usage statistics

This Actor creates a small, content-free summary at the end of each run. It is used only to monitor reliability and improve this Actor. A copy is saved as USAGE_STATS in your own Apify key-value store, so you can see the exact record created for your run.

Set disableUsageStats to true in the input to opt out. Nothing is sent then; your USAGE_STATS record only says that statistics were disabled.

Only these fields are recorded:

• schema version, Actor name and build number;
• UTC start and finish hour (not a precise timestamp);
• run duration, number of results and time to the first result, each as a coarse range;
• whether the result was empty, the end status, and an error type from a fixed list;
• memory setting and counts of charged events;
• names of the input fields you set, never their values;
• the selected option for input fields that offer a fixed list of choices (for example a sort order).

We do not collect input text, search terms, URLs, domains, usernames, email addresses, names, proxy credentials, tokens, scraped records, output items, raw error messages, stack traces, or hashes of any of those values. Records are kept for no longer than 13 months, used only as aggregated operational statistics, and never sold or shared.

Additional fields (Phase 2)

This Actor also records your Apify user ID, whether Apify marks the account as paying, the size range of list inputs, the selected country when the input offers a fixed list of countries, and one category from a fixed Actor taxonomy. We use these fields only for aggregate reliability, repeat-use and cross-Actor analysis; reports suppress any cell with fewer than five distinct users.

The same disableUsageStats: true input flag turns these fields off too. The user ID is removed after 13 months; we do not export, sell, share, or attempt to re-identify this data.

Run-outcome signals (v2)

To learn whether a run did what it was asked to do, the record also holds a few more coarse ranges and yes/no flags. None of them contains content:

• the result limit you asked for (a range, when the input has one) and what share of it was delivered;
• results delivered per input item you listed (a range);
• output quality as ranges: how fully the result fields were filled, the share of rows that look like errors, the share of duplicate rows, and how many different fields appeared. These are counted in memory while results are saved; no result content is kept;
• how the run was started (console, API, schedule, webhook, another Actor);
• how it ended: stopped by you, timed out, reached the requested limit, stopped by the charge limit, and how many times the platform moved the run;
• if this Actor reports it: how many items to process worked or failed (ranges) and one failure reason from a fixed list;
• a short code made from the names of the input fields you set, never their values.

Repeat-run fingerprint (v2)

When your Apify user ID is recorded (see above), the record also holds an 8-character one-way code made from your input (proxy settings left out) and this Actor's name. It only lets us see that the same account ran the same input again soon after an unsatisfying run; we never see the input itself. It is stored only in the database, never published, and reports use it in aggregate with the same five-user minimum. It is the one exception to the statement above that no hashes are collected, and disableUsageStats: true turns it off.