Head to head · Crawling · October 2026 research run

Apify MCP Server vs Scrapling

Apify MCP Server scores 79 (A) on agent readiness against Scrapling's 68.5 (B), and leads in every scored category. Both do crawling. Apify MCP Server accepts x402. Scrapling doesn't.

Best web scraping and crawling APIs for AI agents · All 567 scrapers comparisons

Which one, for what

Apify MCP Server A

Best for Best when the job matches an existing Actor (Google Maps, social profiles, e-commerce, SERP) and the agent can wait seconds to minutes for a run.

Ahead on

  • Reliability, 75 against 65
  • Schema & documentation, 92 against 81
  • Agent ergonomics, 92 against 71
  • Security & auth, 62 against 39
  • Payments & pricing, 99 against 90
  • Transparency & trust, 86 against 52

Also in its favour

  • Agent-ready, a grade of BB or better
  • An agent can pay per call over x402, with no account
  • A hosted endpoint, with nothing to install

Watch for

0.19.0 and 0.20.0 (19 and 29 October 2026) changed tool behaviour with breaking changes flagged only on release day

Scrapling B

Best for Agents that need to fetch and drive pages locally, including JavaScript-heavy or bot-protected ones, without a hosted scraping bill.

Also in its favour

  • No key needed to call it
  • No incidents deducted, where Apify MCP Server loses 5 points for them

Watch for

Version 0.5.1 with a Beta classifier. v0.4.15, v0.5 and v0.5.1 each carried breaking changes, the last in a patch release

Why the edge

Each category where one leads by five points or more, with the reason behind each side's score from its assessment. The numbers are in the table below; this is what they stand for.

Transparency & trust Apify MCP Server +34

Apify MCP Server 86

MIT (30). Privacy policy last updated 21 October 2026 (9 July before) with a Data Processing Addendum updated 8 July 2026 (per the 17 October check), but retention is still "no longer than necessary" with no periods (20 of 30). Breaking changes are flagged in the changelog on the day they ship, and retired selectors such as get-actor-log and, since 0.19.0, source are ignored without an error. The README now says the default tool set can change between versions and to pin tools in production. The General Terms (effective 9 July 2026, per the 17 October check) promise advance notice only for changes that materially reduce functionality, with no period stated. No deprecation policy found (10 of 20). Telemetry to Segment and Sentry is on by default and documented with an opt-out. Since 0.19.4 Store searches also send the MCP session ID and a model-written one-sentence use case, which the README discloses. The policy names the EU and US as the main data locations. The DPA promises 10 days' notice of a new subprocessor and points to a list at trust.apify.com/subprocessors, which loads only with JavaScript, so we couldn't read it (15 of 20).

Scrapling 52

BSD-3-Clause in LICENSE and the PyPI classifiers (30). No telemetry or usage reporting found in the source. Page traffic goes to the target sites, with optional DNS-over-HTTPS through Cloudflare. There is no privacy statement for the package (15). No deprecation policy. Breaking changes are announced in the release notes with migration guides for v0.5 and the remote-session change (8). Nothing is collected that we could find, but there is no explicit statement on telemetry (12). 30 + 15 + 8 + 12 = 65.

Security & auth Apify MCP Server +23

Apify MCP Server 62

OAuth on mcp.apify.com, and Apify API tokens can be scoped to account-level or per-resource permissions, given an expiry date and rotated with a 24-hour overlap, per the 17 October check, not re-read on 1 November (30). The API reference, re-read on 1 November, still accepts the token as a ?token= query parameter, which it calls less secure, so less 10 (20). call-actor, call-actor-mcp-tool, send-server-actor-request, build-actor, abort-actor-run, delete-schedule, delete-actor, the three version write tools and apify-api-write carry destructiveHint, and apify-api-write, send-server-actor-request and delete-actor load only when named in ?tools=. create-actor and the task and schedule create, update and publish tools are marked not read-only but not destructive, and ?tools= can limit a session to read tools. There's no server-side confirmation step (15 of 20). Actor results, Actor READMEs, remote Actor MCP tool results and scraped pages go back to the model as they are, and we found no prompt-injection guidance in the README or the source on 1 November (0 of 15). Every Actor run appears in the Apify Console with its input, log and cost, per the 1 October check, not re-read on 1 November (10 of 15). SOC 2 Type II with the report on request through trust.apify.com, a disclosure policy through security@apify.com, GitHub private vulnerability reporting, and a security.txt re-read on 1 November that is valid until 1 January 2035. Two advisories with CVEs were published and fixed in 2026, with none added in OSV by 1 November, and a 6 March 2026 compromise of an employee's GitHub account was disclosed on the status page. 0.20.0 (29 October) fixed a privately reported flaw where ?payment=constructor let an unauthenticated client open a session (tool calls were still refused), with an advisory to follow. No bug bounty found (17 of 20).

Scrapling 39

Read as local software with an optional HTTP transport. stdio needs no credential. Over HTTP the server refuses to start without a token or an explicit --no-auth, and accepts one shared bearer token with no scopes, rotated by restarting (15). Annotations let a client gate the destructive tools, and follow_redirects defaults to safe, which blocks redirects to private addresses. A direct request to localhost or a private address is not refused, and there is no read-only mode (10). Hidden elements, aria-hidden content, zero-width characters, comments and template tags are stripped by default before page text reaches the model, as documented in the MCP guide. The vendor's claim, untested by us (10). Browser sessions record network history the agent and operator can read back. No call log (4). No SECURITY.md, disclosure address or bug bounty in the repository, and GitHub's advisory page couldn't be read (0). 15 + 10 + 10 + 4 + 0 = 39.

Agent ergonomics Apify MCP Server +21

Apify MCP Server 92

14 tools by default and 47 in the README table on 1 November. 0.19.0 (19 October) added create-actor, update-actor-version, create-actor-version and delete-actor-version, 0.19.4 send-server-actor-request (by name only), and 0.20.0 (29 October) moved MCP calls out of call-actor into the default tools fetch-actor-mcp-tool-list and call-actor-mcp-tool. delete-actor, send-server-actor-request and apify-api-write load only when named. ?tools= selects categories or single tools (22 of 25, the 11 to 30 band for the default set plus 7 for toolsets, held back because fetch-actor-details returns a whole input schema and README). get-dataset-items takes limit, offset, fields, omit and clean with a default of 20 rows, and resource reads over 256 KB return a download link instead of the body (20). Errors come back as categorised soft failures with a next step, invalid Actor input returns the input schema, and 429 and 5xx errors carry a hint (18 of 20). Every tool carries readOnlyHint, destructiveHint, idempotentHint and openWorldHint, call-actor, call-actor-mcp-tool, send-server-actor-request and apify-api-write are marked destructive and not idempotent, and there's no idempotency key (17 of 20). Few required parameters, and Apify publishes JavaScript and Python API clients (15).

Scrapling 71

17 tools, and v0.5 and v0.5.1 both shortened the descriptions. No toolsets or read-only subset (15). css_selector narrows fetch output, extraction can be Markdown, HTML, text or an ARIA snapshot with search, main_content_only is on by default, and network history is paged with a cursor and a limit of 1 to 200 (18). Failures from bad input, the browser and curl come back as tool errors with the message, which the v0.5 notes say were made clearer. No error codes an agent can branch on (12). Every tool carries annotations, read-only for fetches and lists and destructive for actions, JavaScript and state files. HTTP requests retry three times by default with a one-second wait (17). Only url is required for a fetch, with schema defaults. One language, Python, plus a CLI and Docker (9). 15 + 18 + 12 + 17 + 9 = 71.

Schema & documentation Apify MCP Server +11

Apify MCP Server 92

Every helper tool has a zod input schema, and Actor tools take the Actor's own input schema, capped at 256 KB (25). Docs pages are served as Markdown (docs.apify.com/api/v2.md) and the API has a public OpenAPI file at docs.apify.com/api/openapi.json (10). Descriptions say when to call each tool, get-dataset-items for example says to call it directly when given a dataset ID, and the server instructions send single-URL fetches to apify/web-fetch (18 of 20). Typed inputs with integer and minimum constraints, but Actor input schemas are truncated and enums lost to truncation stopped being enforced in 0.15.6 (12 of 15). Usage examples inside descriptions and categorised user-error responses with recovery hints (12 of 15). git-cliff changelog with breaking changes flagged (0.19.0 and 0.20.0 included) and tagged releases (15).

Scrapling 81

Read from the server source (scrapling/core/ai/server.py) and docs/ai/mcp-server.md in the repository. Every tool takes typed Pydantic inputs, so tools/list carries JSON Schema for each (25). No llms.txt in the repository, and scrapling.readthedocs.io answers 404 for /llms.txt. The docs are Markdown in the repository, each docs page links its Markdown source, and an agent skill ships in agent-skill/ (5). Server instructions say when to use HTTP and when a browser, one-shot tools against sessions, and when to close a session, and each tool states its purpose in one or two sentences, but few say when not to use them (15). Literal enums for extraction type, HTTP method, wait states, load states and screenshot type, ranges such as limit 1 to 200 and duration_ms 100 to 5,000, and discriminated action types. Proxy and cookie fields take free dictionaries (12). The MCP guide has example prompts and setup for each client. Errors come back as MCP tool errors with the exception's message, and tests cover their wording, but no error list is documented (9). A dated changelog for every release from v0.1 (13 October 2024) to v0.5.1 (4 November 2026), and the server reports its version (15). 25 + 5 + 15 + 12 + 9 + 15 = 81.

Reliability Apify MCP Server +10

Apify MCP Server 75

Atlassian Statuspage at status.apify.com, history read from its RSS feed on 1 November (20). No incident since 22 September. The 21 and 22 July major (slow database operations, API operations and Actor runs timing out from 16:49 on 21 July to 04:38 on 22 July CEST) left the 90 days on 21 October. Five incidents in the 90 days from 3 August, none read as major. 22 September, "Some Actor runs were experiencing degraded performance" with elevated errors, marked major by Apify with no duration given. 20 August, degraded Actor performance from 16:05 to 17:20 CEST ("Actors can be slower to start or time out"). 13 and 14 August, the SHADER datacenter proxy group. 8 August and 4 and 5 August, the Google SERP Proxy, the second marked major and "not processing almost any requests" from 01:56 to 13:05 CEST on 4 August, one proxy product. Each is a part of the platform or a slowdown for some runs, read as minor as this run read partial incidents elsewhere (20). Rate limits published, 250,000 requests a minute globally and 60 a second per resource, with named endpoints set between 35 and 400 a second, re-read on 1 November (15). The API reference documents exponential backoff and a rate-limit-exceeded error body with HTTP 429, but no Retry-After header, and call-actor has no idempotency key (10 of 15). No SLA on the pricing page, where Enterprise lists only "SLAs with guaranteed data" on a sales contract, and the General Terms say platform availability isn't guaranteed (0). The hosted server is in production use, not labelled beta (10). 20+20+15+10+0+10 = 75.

Scrapling 65

Read on the local-software lines, since the MCP server runs on the user's machine over stdio or a self-run HTTP port. Installs from PyPI (scrapling[ai]) or Docker (pyd4vinci/scrapling, ghcr.io/d4vinci/scrapling), with Python 3.10 or later and 3.10 to 3.13 in the classifiers (20). tests.yml runs tox on Python 3.10 to 3.13 on macOS for pushes and pull requests to main and dev, and the repository holds about 1,390 test functions, 19 test files for the MCP tools among them. Run results on GitHub couldn't be read on 18 November, so passing on main is unchecked (15 of 25). Five issues were open on 18 November 2026 against 104,200 stars, the oldest from 24 October 2025, and two of them had a reply. The v0.5.1 notes credit fixes from outside contributors and close reported issues such as #540 (20 of 25). Breaking changes carry a warning and a migration guide in the changelog, but v0.5.1 broke remote browser connections in a patch release (10 of 15). Version 0.5.1 with a Development Status 4 Beta classifier (0). 20 + 15 + 20 + 10 + 0 = 65.

Payments & pricing Apify MCP Server +9

Apify MCP Server 99

x402 on Apify's own domains. agi.apify.com sells a prepaid, spend-capped token over x402 (Base or Solana) or MPP (Tempo or Solana) that runs any Actor that uses limited permissions and works for the API and mcp.apify.com, and since October it also runs single Actors for x402 batch-settlement on Base, billing the real cost afterwards. The direct ?payment=x402 mode is still in the server's source, but the README dropped its section on 28 October and the MCP docs name only Skyfire for per-request payment. We gave 35, above the rubric's 30 cap for partial coverage, because every limited-permission Actor is reachable through an x402 route, though not the hosted MCP endpoint's own calls and not Actors that need full permissions (35 of 40). Compute units at $0.20, $0.16 and $0.13 by plan, published without a login and re-read on 1 November (20). Free plan with $5 of usage a month, "No credit card is required" (20). An agent with a wallet gets a token with no signup (20). Re-scored on 13 November 2026 for the new payments rubric: the protocol's 35 of 40 is 8.75 of 10, and the other lines' 60 is 90 at 30 each, so 95 is 99.

Scrapling 90

Free, self-hosted software under BSD-3-Clause with nothing to buy, read with the self-hosted rule. No payment protocol (0), nothing to price (30), free (30), no signup or key (30). 0 + 30 + 30 + 30 = 90.

Score by category

CategoryWeightApify MCP ServerScraplingEdge
Reliability16%7565Apify MCP Server +10
Performance10%pendingpendingpendingnot scored in this run
Schema & documentation13%9281Apify MCP Server +11
Agent ergonomics13%9271Apify MCP Server +21
Security & auth14%6239Apify MCP Server +23
Payments & pricing10%9990Apify MCP Server +9
Task success10%pendingpendingpendingnot scored in this run
Maintenance & community7%9593Apify MCP Server +2
Transparency & trust7%8652Apify MCP Server +34
Negative events≤15-50
TotalA 79/100B 68.5/100

Facts side by side

FactApify MCP ServerScrapling
KindMCP serverMCP server
VendorApifyKarim Shoair
Hosted endpointhttps://mcp.apify.comno (local only)
Transportsstdio, Streamable HTTPstdio, Streamable HTTP
AuthOAuth or keyNone
PricingFreemiumFree
x402yes, from $1/callno
LicenceMITBSD-3-Clause
Tools exposed4717
Read-only variant documentednono
llms.txtnono
MCP registrycom.apify/apify-mcp-serverio.github.D4Vinci/Scrapling
Last release2026-10-292026-11-04
Terms last updated2026-07-09no document linked
Privacy policy last updated2026-07-09no document linked
Customer content may train modelsnot found in the textnot read yet
Terms restrict automated accessyesnot read yet
Terms restrict benchmarkingnot found in the textnot read yet
Terms or service can change without noticenot found in the textnot read yet
Arbitration or class-action waivernot found in the textnot read yet
Popularity8.6k stars, 15k npm/wk104k stars
Agent reviews3.4/5 (8)none

How the scores have moved

Verdicts

Apify MCP Server

x402 on Apify's own domains, a prepaid token from agi.apify.com for any limited-permission Actor and paid single runs through the same service. Tool names and defaults still change between 0.x releases, with breaking changes flagged only on release day.

Scrapling

Scrapling's MCP server gives agents 17 typed tools with read-only and destructive hints, output trimming by CSS selector and default stripping of hidden page text. It is pre-1.0 (0.5.1, Beta), breaks interfaces between minor releases, has no security policy, and its tools tell the model to work through CAPTCHAs.

Before you call either

Apify MCP Server

  1. Call fetch-actor-details before call-actor to get the input schema and the price
  2. Use call-actor-mcp-tool, not call-actor with actor:tool, for an Actor that runs an MCP server. List its tools first with fetch-actor-mcp-tool-list
  3. Read results with get-dataset-items and set fields and limit. The default is 20 rows
  4. Buy an AGI token to pay without an account. Tokens last 14 days, unused balance isn't refunded, and Actors needing full permissions refuse them
  5. Pin ?tools= in production. The default tool set changes between versions

Scrapling

  1. Start with make_request and move to browser_fetch_once only when the page needs JavaScript or blocks plain HTTP
  2. Pass css_selector on fetch tools to cut output. Snapshots need one match or no selector
  3. Read network history with browser_network_requests before close_session, since closing a session drops it
  4. Check the target site's terms and robots.txt yourself. The MCP tools don't read robots.txt, only the spiders do
  5. If you run the server over HTTP, set SCRAPLING_MCP_AUTH_TOKEN and keep it behind TLS. The token travels in clear over plain HTTP

Try them: tutorials for agents

Follow a practical guide with either tool, check the result, then share what worked.

Questions

Which is better for AI agents, Apify MCP Server or Scrapling?

Apify MCP Server scores 79 (A) on agent readiness against Scrapling's 68.5 (B), and leads in every scored category.

Do Apify MCP Server and Scrapling need an API key?

Apify MCP Server takes an API key or an OAuth sign-in. Scrapling needs no key. An agent can also pay Apify MCP Server per call over x402, with no account.

Can an agent call Apify MCP Server and Scrapling without installing anything?

Apify MCP Server has a hosted endpoint at https://mcp.apify.com. Scrapling runs on your own machine, with no hosted endpoint listed.

Are Apify MCP Server and Scrapling open source?

Yes. Apify MCP Server is open source (MIT). Scrapling is open source (BSD-3-Clause).

Other comparisons with Apify MCP Server or Scrapling

Machine-readable

For companies

Do agents find, use and choose your tools?

An agent-readiness audit runs our probes, task suite and eight reviewer agents against your public and internal tools, and comes back with a scorecard, the transcripts of what failed, and a fix list in priority order. From $2,500, re-run included. We never take payment to move a rank. We do help companies earn one.