Issue 02AgentX field manual2026-09-04
The field manual for working web data.
Choose the source. Understand the contract. Run a narrow sample. BestCrawler maps every public AgentX and TrueFetch Actor and publishes the practical guide that belongs beside it.
Choose a desk
01—05Start with the question,
not the scraper.
Five topic desks connect a real research or automation need to the right current AgentX route.
Job data
Guides to multi-source job collection, hiring signals, salary research, and company enrichment, plus the pipelines that keep job data usable in production.
02Read desk ↗Social intelligence
Methods for public community, creator, trend, post, and audience research across Telegram, Reddit, Instagram, and TikTok, with the source caveats that matter.
03Read desk ↗Video knowledge
Working with transcripts, captions, translations, and downloads from public video, and turning them into knowledge pipelines you can search and reuse.
04Read desk ↗Market signals
Commerce, property, vehicle, and local-business intelligence: how to collect pricing, inventory, and listing data across markets without over-reading it.
05Read desk ↗Data operations
The operational half of web data: choosing Actors, calling APIs, validating datasets, managing crawler access, and delivering results reliably in production.
The directory
Current / 2026-09-04Six useful ways in.
Popularity is a discovery signal, not a quality verdict. Open the live contract and test the output that matters to your workflow.
Browse all 152 Actors →All Jobs Scraper (39 Sites)
All Jobs Scraper is a job scraper API for 42 platforms that returns 41 unified job, salary, company, application, and public contact fields. One country-aware input can search multiple job boards in the same run.
Telegram Channel Message
Telegram channel scraper for dependable post-history exports, returning 55 structured fields plus optional stored media. Capture text, senders, dates, views, replies, forwards, reactions, topics, service events, links, hashtags, and media from any public username.
Instagram Trending Scraper
Unofficial Instagram Explore scraper for public posts and reels by country, with section and topic labels, captions, engagement metrics, creator, location, comments, audio, and automatically archived images and videos. Not affiliated with Instagram or Meta.
Telegram Profile
Telegram profile scraper for resolving up to 1,000 usernames, @handles, t.me URLs, accessible invites, or phone numbers into 41-field sparse profiles for users, bots, channels, groups, and supergroups, including identity, public flags, counts, business details, and photo links.
Instagram Video Transcript
Instagram Video Transcript extracts the spoken words from one public Reel or video post for hook research, captions, and content repurposing. It returns detected-language text, ordered timestamps, post metadata, and optional translation into 133 languages. Entry transcript price: $0.36.
Telegram Member Scraper
Telegram member scraper for public groups, returning 23-field profiles from the visible participant list and optional message-author discovery. Export IDs, usernames, names, public flags, presence, bot facts, discovery path, and source-group context.
Order uses public Apify Store 30-day user totals captured 2026-09-04. The linked Actor page remains authoritative.
The journal
34 guides / 5 desksNo-fluff guides for the work after “scrape it.”
Selection, normalization, quality, evidence, delivery, and the limits that keep a dataset honest.
Open the full journal →Field note 05 · 8 min read
How to Give an AI Agent Live Web Data
A practical guide to wiring real-time web data into AI agents: choosing tools over scrapers, designing the data contract an LLM can reason about, and controlling cost and failure.
Operations guide 07 · 7 min read
Google Trends Data as an API: Programmatic Keyword Research Without the Export Button
How to turn Google Trends into a repeatable data feed for SEO and market research: batching keyword comparisons, reading interest timelines, and mining rising queries at scale.
Jobs guide 01 · 9 min read
How to Collect Job Listings From Multiple Sites in One Dataset
A practical guide to multi-source job data: when to use an orchestrator, which fields to normalize, how to deduplicate listings, and how to verify coverage.
Operations guide 05 · 11 min read
How to Run an Apify Actor by API in a Production Application
A production integration pattern for Actor input, asynchronous runs, status handling, datasets, secrets, retries, idempotency, and observable delivery.
Jobs guide 04 · 11 min read
How to Build a Hiring Intelligence Pipeline With Public Job Data
A production blueprint for turning public job listings into hiring signals: collection, raw storage, normalization, deduplication, enrichment, QA, and alerts.
Field note 04 · 8 min read
How to Choose an Apify Actor for a Production Data Workflow
A source-first method for comparing Apify Actors by entity, geography, schema, freshness, pricing, limits, and a real sample run before scaling.
The four-step test
A tool is only useful when the contract survives contact with real data.
Open the playbooks →Name the entity
Job, company, profile, post, video, product, place, property, or vehicle—the contract starts with the row.
Choose the source
Use a multi-source route for one schema; use a dedicated Actor when a source-specific field is decisive.
Prove a sample
Inspect source URLs, missing fields, dates, units, duplicates, and failure behavior before increasing volume.
Keep the evidence
Raw rows, run metadata, source links, and transformation versions stay attached to every conclusion.
Desk notes
Before you run.
What is BestCrawler?
BestCrawler is the editorial discovery layer and field manual for the AgentX and TrueFetch public Apify Actors. It helps a team choose a data route, validate the contract, and open the current Actor on Apify.
How many Actors are indexed?
The directory contains 152 public AgentX and TrueFetch Actors returned by the Apify Store API in the 2026-09-04 catalog snapshot.
Are the guides generated product descriptions?
No. The journal is organized around real workflow questions, source evidence, data contracts, quality checks, and explicit limitations. Product-specific behavior is always verified on the linked live Actor page.
Where should a new user start?
Start with one of the five topic desks, then compare the recommended Actors and run a narrow sample using the current input and output contract on Apify.