Stack Overflow questions + answers search — Stack Exchange API 2.3

One row per Stack Overflow question matched by your query, tags, sort and date range — question id, title, tags, score, view count, answer count, answered / accepted-answer flags, creation and last-activity dates, owner display name and reputation, canonical link and the body as plain text — followed by the question's top answers as rows of row_type = answer. Works for any Stack Exchange site (site input). Read through the public Stack Exchange API 2.3 (https://api.stackexchange.com/docs), no login.

**The default input (query asyncio, tag python, last activity first, 100 questions

• up to 3 answers each) succeeds within 5 minutes and returns rows** — Apify auto-tests it daily.

Data source, license and status

Publisher	Stack Exchange, Inc. (Stack Overflow and the Stack Exchange network)
API	https://api.stackexchange.com/2.3 — public, documented at https://api.stackexchange.com/docs; /search/advanced for questions, /questions/{ids}/answers for answers, both with filter=withbody
Quota	300 requests per day per IP without a key. Your own app key (stackexchange_key, registered free at https://stackapps.com/apps/oauth/register) raises it to 10,000. Each 100 questions cost one request; answers cost about one request per 30 questions
Rate	At most one request per second by default; the API's backoff field is honoured; throttle errors (error_id 502) are retried after 30 s
License / attribution	User contributions on Stack Exchange sites are licensed under Creative Commons Attribution-ShareAlike (CC BY-SA, version per post: https://stackoverflow.com/help/licensing). Every row carries the API's own content_license as license, the author as owner_display_name and the canonical link. If you republish the text you must attribute the author, link to the post, and share alike
API terms	https://stackoverflow.com/legal/api-terms-of-use — this Actor identifies itself, respects the quota and backoff, and caches nothing
Known caveats	body_text is the HTML body with tags stripped and cut at 3,000 characters (code blocks lose their formatting); view_count, answer_count, is_answered are null on answer rows; accepted_answer_id is absent when nothing was accepted; deleted or closed questions are whatever the API returns

Honesty note: the rows are the API's own fields renamed and flattened — nothing is inferred, summarised or classified by this Actor, and a run stops with an error rather than guess when the API's shape changes.

Identification: the requests carry a product-token User-Agent and the operator's contact address in the standard From: header (RFC 9110 §10.1.2).

Input

Field	Type	Meaning
query	string	Free-text query (q); optional when tags are given
tags	array	Up to 5 tags, all must match (tagged=a;b)
sort	activity / votes / creation	Default activity, always descending
from_date / to_date	YYYY-MM-DD	Optional creation-date bounds (UTC, inclusive)
include_answers	boolean	Default true — top answers (by votes) as extra rows
answers_per_question	integer	Default 3, at most 20
max_records	integer	Maximum question rows; default 100, at most 5,000
site	string	Default stackoverflow; any site from https://api.stackexchange.com/docs/sites
stackexchange_key	secret string	Optional app key, raises the daily quota
request_interval_ms	integer	Default 1000 (floor 250)
contact_email	string	Sent in the From: header

Example — highest-voted pandas questions about groupby created in 2026, no answers:

{ "query": "groupby", "tags": ["python", "pandas"], "sort": "votes", "from_date": "2026-01-01", "include_answers": false, "max_records": 50 }

Output (dataset row)

row_type (question / answer), question_id, answer_id, site, title, tags, score, view_count, answer_count, is_answered, accepted_answer_id, is_accepted, creation_date, last_activity_date (ISO 8601 UTC), owner_display_name, owner_reputation, owner_user_id, link, body_text (≤ 3,000 chars), license, fetched_at, source.

Answer rows repeat the parent question's title, tags and accepted_answer_id so each row stands alone; their link is the question link with #answer_id.

A run summary (filters, row counts, requests, remaining daily quota, whether more pages were available or the pay-per-event budget stopped the run) is stored as RUN_SUMMARY in the run's key-value store.

Pricing (pay per event)

Event	Price
run-start — once per run	$0.10
record — per row written (question or answer)	$0.005

The default pull (100 questions + up to 300 answers) costs at most $2.10; 100 questions without answers cost $0.60. Rows stop when your run's maximum charge is reached; the run summary says so.

Operator

Steelyard Ventures LLC (RetrainMap) — info@steelyardclinical.com. Not affiliated with Stack Exchange, Inc. The Actor writes only to its own dataset and key-value store; the optional API key is sent only to api.stackexchange.com and never stored.