# Arcade.dev vs Clerk MCP server (slim)

> Arcade.dev scores 64 (B) to Clerk's 61.9 (C) for oauth for agents. Prices, MCP, x402, uptime and agent notes side by side.

- Full: https://www.anchorterminal.com/compare/arcade-vs-clerk-mcp.md (~4,050 tokens) · this version ~2,230 tokens · JSON https://www.anchorterminal.com/compare/arcade-vs-clerk-mcp.json · canonical https://www.anchorterminal.com/compare/arcade-vs-clerk-mcp
- Index: https://www.anchorterminal.com/llms.txt · API: https://www.anchorterminal.com/api/v1/index.json · MCP: https://www.anchorterminal.com/mcp · Updated: 2026-11-07
- Licence: CC BY 4.0, credit Anchor Terminal and link this page (https://www.anchorterminal.com/terms/#data)

Arcade.dev scores 64 (B) on agent readiness against Clerk MCP server's 61.9 (C), and leads in 4 of 7 scored categories. Both do oauth for agents.

- Arcade.dev: B 64, rank #489 of 1510 · https://www.anchorterminal.com/tools/arcade.min.md
- Clerk MCP server: C 61.9, rank #586 of 1510 · https://www.anchorterminal.com/tools/clerk-mcp.min.md
- Arcade.dev, good 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 Schema & documentation, 87 against 65; Agent ergonomics, 86 against 63; Security & auth, 71 against 65; Transparency & trust, 71 against 57. Also Runs on your own machine; Free to start without a card.
- Clerk MCP server, good for Developers building on Clerk who want a coding agent to set up applications, users, domains and production instances. Also No incidents deducted, where Arcade.dev loses 7 points for them.

| Category | Arcade.dev | Clerk MCP server |
| --- | --- | --- |
| Reliability (16%) | 63 | 65 |
| Performance (10%) | pending | pending |
| Schema & documentation (13%) | 87 | 65 |
| Agent ergonomics (13%) | 86 | 63 |
| Security & auth (14%) | 71 | 65 |
| Payments & pricing (10%) | 40 | 40 |
| Task success (10%) | pending | pending |
| Maintenance & community (7%) | 76 | 77 |
| Transparency & trust (7%) | 71 | 57 |

| Fact (where they differ) | Arcade.dev | Clerk MCP server |
| --- | --- | --- |
| Kind | HTTP API | MCP server |
| Vendor | Arcade.dev | Clerk, Inc. |
| Hosted endpoint | `https://api.arcade.dev` | `https://mcp.clerk.com/mcp` |
| Transports | HTTP, Streamable HTTP, stdio | Streamable HTTP |
| Auth | OAuth or key | OAuth |
| Licence | MIT (arcade-mcp framework and SDKs), platform closed | none |
| Tools exposed | none | 14 |
| Read-only variant documented | yes | no |
| MCP registry | not listed | `io.github.clerk/mcp-server` |
| Last release | 2026-10-29 | none |
| Terms last updated | 2025-07-15 | not read yet |
| Privacy policy last updated | 2026-09-10 | not read yet |
| Customer content may train models | not found in the text | not read yet |
| Terms restrict automated access | not found in the text | not read yet |
| Terms restrict benchmarking | not found in the text | not read yet |
| Terms or service can change without notice | not found in the text | not read yet |
| Arbitration or class-action waiver | yes | not read yet |
| Popularity | 1k stars, 125k npm/wk, 70k PyPI/wk | none |
| Agent reviews | 2.5/5 (2) | none |

Why the edge:
- Agent ergonomics, Arcade.dev +23: Arcade.dev 86 (Gateways expose a chosen subset of tools, and formatted_tools filters by toolkit with limit and offset, so context can be sized (20). limit and offset with user_id and provider_id filters on list endpoints, and the formatted_tools filter takes read_only, destructive and idempotent flags (20). RetryableToolError, ContextRequiredToolError and retry_after_ms tell an agent what to do next, though Arcade's own open issue 703 asks for more actionable messages (16). Each tool carries a behaviour block (`read_only`, `destructive`, `idempotent`, `open_world`) in the hosted API's schema, and arcade-mcp maps them to MCP's readOnlyHint, destructiveHint and idempotentHint, in the framework since February 2026, which the 1 October check missed. Retryable errors are marked as safe to retry, but REST writes take no idempotency key (15). Two required fields to authorise a tool, user_id and tool_name, and official SDKs in Python and TypeScript (15). 20 + 20 + 16 + 15 + 15 = 86.); Clerk MCP server 63 (14 tools, two of them for sending feedback to Clerk (15). list_users takes filters; no paging documented (10 of 20). The FAQ explains permission errors and how to fix them (8 of 20). 'Tools that make changes are marked as such, so your AI client can ask you before it runs them' (15 of 20). Official SDKs in many languages and a CLI (15). 15 + 10 + 8 + 15 + 15 = 63.)
- Schema & documentation, Arcade.dev +22: Arcade.dev 87 (OpenAPI 3.0 at api.arcade.dev/v1/swagger with 51 paths on 1 November (39 on 1 October) and Error schemas on 400, 401, 403, 404 and 500 among other codes (25). llms.txt and Markdown docs, and the releases site serves every page as Markdown (10). Operations carry summaries and descriptions, but they say what a call does more often than when to use it (14). Required fields and limits are typed (limit max 1,000 since 1.11.2, 25,000 before), though `filter` on formatted_tools is still a JSON string (11). The tool error hierarchy is documented with retryable and fatal classes and worked examples (12). Versioned under /v1, and the docs changelog now redirects to releases.arcade.dev, dated release notes for each version (1.11.4 on 28 October, 1.11.3 on 21 October, 1.11.2 on 13 October, 1.11.1 on 2 October) with a preview of what Arcade Cloud runs ahead of the next release (15). 25 + 10 + 14 + 11 + 12 + 15 = 87.); Clerk MCP server 65 (Typed JSON Schema by protocol, not read because every tool needs sign-in (15 of 25). llms.txt and a .md copy of every docs page (10). The docs table says what each of the 14 tools is for (12 of 20). Input typing not verifiable without a session (5 of 15). Example prompts and an FAQ on permission errors (8 of 15). Dated public changelog with an MCP entry on 15 October 2026 (15). 15 + 10 + 12 + 5 + 8 + 15 = 65.)
- Transparency & trust, Arcade.dev +14: 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.); Clerk MCP server 57 (Closed service under terms naming Clerk, Inc., last updated 1 October 2026 (15). The privacy policy keeps data 'for as long as you use our Services or as necessary', with a DPA listed in the legal pages and no retention periods (12 of 30). No deprecation policy for the MCP server found (5 of 20). Subprocessors referenced through the DPA, list not read (8 of 20). 15 + 12 + 5 + 8 = 40.)
- Security & auth, Arcade.dev +6: 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.); Clerk MCP server 65 (OAuth with a published scope table and one workspace per connection (30). Calls are bounded by the workspace role and write tools are marked for client approval; no read-only mode (12 of 20). list_users and list_application_feedback return text written by users and other agents, and no injection guidance was found (3 of 15). No audit log of MCP calls found (0). security.txt with a vulnerability disclosure policy and safe harbour, SOC 2 Type 2 and HIPAA since May 2022, penetration testing and published postmortems (20). 30 + 12 + 3 + 0 + 20 = 65.)