Harness Hub
All startups

Arcade

Authorization and tool infrastructure for AI agents.

Arcade (arcade.dev) builds the identity and authorization layer for agents — OAuth-backed tool execution, credential management and fine-grained access control so agents can act on a user's behalf without inheriting god-mode credentials. Raised a $60M Series A in 2026.

Pros & cons

Each point cites its sources and says what kind of evidence backs it. A second reviewer tried to refute every claim before it was kept.

Pros

  • DocsHandles user-scoped OAuth for agent tool calls, so agents act as the end user instead of through over-scoped service accounts.[1][2]
  • DocsShips a large catalogue of agent-optimised tools (about 4,000 across Gmail, Slack, GitHub, Salesforce and more).[1]
  • MetadataIts arcade-mcp framework for building MCP servers with auth is MIT-licensed.[1]
  • DocsCo-authored the MCP URL-elicitation proposal for external auth with Anthropic, so it is closely tied to the spec.[1]

Cons

  • Independent testThe open-source part is the framework; the hosted runtime, managed auth and enterprise controls are commercial.[1][2]
  • User reportsA G2 reviewer said control over auth provider setup is limited.[1]
  • Independent testUsage-based pricing on tool executions and hosted server hours makes costs scale with agent traffic, and enterprise tiers are custom-quoted.[1]
  • Independent testArcade's own test of Anthropic's tool search over its ~4,000 tools found retrieval much harder than with small toolsets, a limit when exposing big catalogues.[1]
Best for
Teams building multi-user agents that need to act inside users' SaaS accounts with proper OAuth.
Watch out for
Dependence on the commercial runtime for managed auth, and usage-based costs.

Assessed Oct 29, 2026 · adversarially reviewedSuggest a correction

People

On X

No posts from them in our X feed yet.

Follow @TryArcade on X · The X feed

In the news

Arcade.dev raises a $60M Series A for its agent authorization layer — OAuth-backed tool execution and credential management so agents act on users' behalf without god-mode access. 'What agents can do' joins execution and governance as a funded category of the harness stack.

The integration layer: hundreds of managed tools your agents can call.

Composio handles auth, schemas, observability and reliability for the external SaaS tools an agent uses — 'plumbing for agentic actions' across Slack, GitHub, Salesforce and 250+ other surfaces.

4 pros · 4 cons

A registry and toolchain for MCP servers.

Smithery indexes and helps deploy MCP (Model Context Protocol) servers — a distribution layer for the tool ecosystem agents connect to.

3 pros · 4 cons

Exa

Search API purpose-built for AI agents.

Exa (formerly Metaphor) provides neural search and web-crawl APIs tuned for agent consumption — live, cited knowledge for agents instead of stale training data.

4 pros · 4 cons

The Harness Hub newsletter

New harness companies, projects, releases and the monthly State of Harness report, in your inbox.