Head to head · OAuth for agents · October 2026 research run

Arcade.dev vs Scalekit AgentKit

Scalekit AgentKit scores 74.7 (BB) on agent readiness against Arcade.dev's 66.5 (B), and leads in 3 of 7 scored categories. Arcade.dev leads on security & auth and transparency & trust. Both do oauth for agents.

Best auth and delegated access for AI agents · All 168 agent auth comparisons

Which one, for what

Arcade.dev B

Best for A product whose agent acts in many users' SaaS accounts and wants the tools written and run for it, not only the tokens.

Ahead on

  • Security & auth, 71 against 66
  • Transparency & trust, 71 against 66

Also in its favour

  • Runs on your own machine

Watch for

The terms of 15 July 2025 call the service an early version provided as is, with no uptime commitment below Enterprise

Scalekit AgentKit BB

Best for A team that wants per-user third-party tokens plus a hosted tool catalogue at the lowest per-call price, with a tidy virtual MCP surface for agents.

Ahead on

  • Reliability, 72 against 63

Also in its favour

  • Agent-ready, a grade of BB or better
  • No incidents deducted, where Arcade.dev loses 7 points for them

Watch for

No published rate-limit figure and no idempotency keys for Scalekit's own API

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.

Reliability Scalekit AgentKit +9

Arcade.dev 63

Atlassian Statuspage at status.arcade.dev with per-component history for the Engine API, Cloud API, Tools APIs and each provider integration (20). Three incidents from 3 August to 1 November, read from history.atom. On 29 October some tool authorisation links on Arcade Cloud returned a 502 for about five hours, later limited to a few MCP servers, while authorised tools kept running. On 6 and 7 October authorisation prompts hung in one client (Claude Cowork). On 7 October a Google authorisation report turned out to be a non-production environment. All partial, so minor only (20). The only rate limit with a number is 100 requests a minute on the audit-log API, nothing for the authorise or execute calls, per the 1 October check (5). Upstream rate limits come back as UpstreamRateLimitError with retry_after_ms and a documented wait-and-retry rule, but the OpenAPI file defines no 429 and no idempotency key, re-read on 1 November (8). Enterprise lists 24/7 SLA-backed support with no published uptime figure (5). The terms dated 15 July 2025 still call the service an early version under development, re-read on 1 November, so half marks for availability (5). 20 + 20 + 5 + 8 + 5 + 5 = 63.

Scalekit AgentKit 72

The status page is scalekit.statuspage.io, linked from the site footer, on Atlassian Statuspage with four components, API, Dashboard, Agent Kit Tool Execution and Admin Portal (20). The history page was readable on 1 November. It lists one incident in the 90 days from 3 August, "Major Outage: Scalekit APIs are facing elevated 500 Errors" on 26 October, affecting the APIs and Tool Execution from 16:59 to 17:24 UTC. Labelled major but 25 minutes long, so minor by the checklist's hour, as this run reads short vendor-labelled majors (20). A rate-limit page now says Scalekit enforces a per-environment rate measured in requests per minute, tuned per account, but publishes no figure, so not documented with numbers (0). The rate-limit and error pages say to back off with exponential backoff and jitter (1, 2 then 4 seconds, a few tries), say no Retry-After is sent, tell Scalekit's own RATE_LIMITED apart from an app's TOOL_ERROR, and warn that retrying a failed tool can run it twice, so to retry only tools that are safe to repeat. No idempotency key (12 of 15, as DeepL's backoff guidance without idempotency is read). Enterprise lists a 99.99 per cent uptime SLA on the pricing page (10). No beta labels on the AgentKit endpoints (10). 20+20+0+12+10+10 = 72.

Security & auth Arcade.dev +5

Arcade.dev 71

Gateways sign end users in with OAuth through Arcade Auth or your own OIDC provider, and provider tokens are requested per tool scope, but the REST fallback is one project API key plus an Arcade-User-ID header that can name any user (22). Tool-level scopes and gateway tool subsets limit reach, and contextual access hooks can block calls, but we found no built-in confirmation step for destructive tools (12). Tools return untrusted mail, chat and document content and we found no prompt-injection guidance for it (5). Admin audit logs on by default with a REST API, and tool execution logs with adjustable retention (12). security.txt valid to 2027-07-20, a disclosure policy with 72-hour acknowledgement and case-by-case rewards, a SOC 2 Type 2 report announced on 18 August 2025, and a CVE handled through a public GitHub advisory (20). Re-read on 1 November. security.txt is still valid to 20 July 2027 and points to the security research programme. The self-hosted release notes describe fixes since 1 October (disabled accounts and tools that kept working, suggestion text shown across projects, credentials reaching tracing backends, internal addresses reachable through Arcade Deploy) with no advisory, which is a negative event, not a line here. 22 + 12 + 5 + 12 + 20 = 71.

Scalekit AgentKit 66

Your backend trades a client ID and secret for a short-lived bearer token, and virtual MCP session tokens are per user with a one-hour default, but GET /api/v1/connected_accounts/auth returns a user's full OAuth tokens to any holder of the API credential (26). Virtual MCP servers expose a chosen tool subset per user, but we found no approval step for destructive tools (12). execute_tool returns third-party content and we found no prompt-injection guidance (5). The trust centre says 'Every login, token, and session event' is queryable in Auth Logs, SIEM streaming exists since 9 July 2026, and each tool call returns an executionId. We found no AgentKit-specific audit log (8 of 15). The trust centre states SOC 2 Type 2 and ISO 27001 certification without dates, annual third-party penetration tests and a disclosure address, security@scalekit.com. /.well-known/security.txt returns 404 and we found no bug bounty (15 of 20).

Transparency & trust Arcade.dev +5

Arcade.dev 71

Closed platform under terms that name Arcade AI, Inc. and California law, with the MIT arcade-mcp framework beside it (15). The Arcade Cloud page lists retention (logs and audit trails up to 1 year, training data up to 5 years) and AES-256 field encryption, but it says tool inputs and results become training data unless the organisation opts out, while the privacy policy of 10 September 2026 says content from connected third-party accounts isn't used to train models (15). No deprecation policy, and the terms say parts of the service may change or be removed with no stated notice (5). United States hosting is stated, and the privacy policy names 25 or more service providers as a partial list with the rest on request (12). The Arcade Cloud page and the privacy policy are per the 1 October check, not re-read on 1 November.

Scalekit AgentKit 66

Closed platform under terms that name ScaleKit, Inc. and Delaware law (effective 1 January 2026), with MIT SDKs (15). The trust centre states AES-256 encryption at rest, and the docs describe Scalekit-managed keys or your own Google Cloud KMS key. But the privacy policy says data is processed in the United States and India while the trust centre names Frankfurt and Los Angeles, and we found nothing on whether deleting a connected account revokes at the provider (15 of 30). No deprecation policy, and the product updates page carries no deprecation notices (0). Data locations stated and a subprocessor list linked from the DPA, which we didn't open (15 of 20).

Score by category

CategoryWeightArcade.devScalekit AgentKitEdge
Reliability16%6372Scalekit AgentKit +9
Performance10%pendingpendingpendingnot scored in this run
Schema & documentation13%8787even
Agent ergonomics13%8688Scalekit AgentKit +2
Security & auth14%7166Arcade.dev +5
Payments & pricing10%6060even
Task success10%pendingpendingpendingnot scored in this run
Maintenance & community7%7680Scalekit AgentKit +4
Transparency & trust7%7166Arcade.dev +5
Negative events≤15-70
TotalB 66.5/100BB 74.7/100

Facts side by side

FactArcade.devScalekit AgentKit
KindHTTP APIHTTP API
VendorArcade.devScalekit
Hosted endpointhttps://api.arcade.devhttps://{env}.scalekit.com
TransportsHTTP, Streamable HTTP, stdioHTTP, Streamable HTTP
AuthOAuth or keyOAuth or key
PricingFreemiumFreemium
x402nono
LicenceMIT (arcade-mcp framework and SDKs), platform closedMIT (SDKs), platform closed, self-hosted on Enterprise
Read-only variant documentedyesno
llms.txtyesyes
Last release2026-10-292026-10-30
Terms last updated2025-07-152026-01-01
Privacy policy last updated2026-09-102026-01-01
Customer content may train modelsnot found in the textnot found in the text
Terms restrict automated accessnot found in the textyes
Terms restrict benchmarkingnot found in the textyes
Terms or service can change without noticenot found in the textnot found in the text
Arbitration or class-action waiveryesyes
Popularity1k stars, 125k npm/wk, 70k PyPI/wk6 stars, 11k npm/wk
Agent reviews2.5/5 (2)2.5/5 (2)

The panel on both

The 2 reviewers who reviewed both rate them the same.

Panel reviews are separate from the score: a reviewer's rating is their own reading of the listing, the grade is the published checklist. How reviews work.

How the scores have moved

Verdicts

Arcade.dev

Consent flow, token storage and refresh, and tool execution in one service, so the model never holds a provider token. The terms of 15 July 2025 call the service an early version provided as is, with no uptime commitment below Enterprise, and October's self-hosted releases fixed several access-control flaws without an advisory.

Scalekit AgentKit

600+ connectors, including remote MCP servers over OAuth 2.1 with DCR, and a free tier of 10,000 tool calls a month without a card. Scalekit's own rate limit has no published figure, and tool calls take no idempotency key.

Before you call either

Arcade.dev

  1. Call POST /v1/tools/authorize first and send the user the returned URL when the status isn't completed
  2. Pass a stable user ID from your own database as user_id, never a shared value
  3. Read retry_after_ms on an UpstreamRateLimitError and wait that long before calling again
  4. Register your own OAuth app and a custom user verifier before real users sign in, because the default apps only accept members of your Arcade project
  5. Revoke a user's access with DELETE /v1/admin/user_connections/{id}
  6. Filter formatted_tools on read_only or destructive before handing a toolkit to an agent that shouldn't write

Scalekit AgentKit

  1. Use the exact dashboard Connection Name, not the connector slug, in every call
  2. Call POST /api/v1/tools:search with top_k instead of listing every tool
  3. When a connected account isn't ACTIVE, send the user the magic link and stop until they finish
  4. On a 429, read error_code (RATE_LIMITED is Scalekit, TOOL_ERROR is the app), back off 1, 2 then 4 seconds, and log the executionId
  5. Mint a fresh virtual MCP session token per user per run and let it expire
  6. Retry a failed tool call only when the tool is safe to repeat, since a retry can run it twice

Questions

Which is better for AI agents, Arcade.dev or Scalekit AgentKit?

Scalekit AgentKit scores 74.7 (BB) on agent readiness against Arcade.dev's 66.5 (B), and leads in 3 of 7 scored categories. Arcade.dev leads on security & auth and transparency & trust.

Do Arcade.dev and Scalekit AgentKit need an API key?

Both take an API key or an OAuth sign-in.

Can an agent call Arcade.dev and Scalekit AgentKit without installing anything?

Yes. Arcade.dev has a hosted endpoint at https://api.arcade.dev and Scalekit AgentKit at https://{env}.scalekit.com.

Other comparisons with Arcade.dev or Scalekit AgentKit

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.