
1 / 3
【Overview】 Corsair is an open source integration layer that lets AI products and agents connect to third-party apps using OAuth, permissions, webhooks, MCP, and typed API operations. It manages multi-tenant credential isolation, so developers don't have to build and maintain per-provider authentication and API infrastructure themselves. With Corsair, you can add secure integrations to your AI agent or product without reinventing the plumbing for every service. Corsair is aimed at AI product teams, developers building AI agents, and teams in production that need multi-tenant integrations. It is not intended for teams that only need plain OAuth authorization, for which the site suggests Nango instead. 【Key features】 • Typed API operations per integration: Every plugin provides a suite of typed operations with compile-time inferred inputs and outputs for calling third-party APIs. • OAuth and multi-tenant credential management: Handles OAuth connect flows and per-tenant credential isolation, resolving credentials internally at call time. • Permission modes and approvals: Set a permission mode per integration so destructive actions require explicit user approval before executing. • Webhooks and triggers: Processes provider webhooks to keep cached data fresh and enable trigger-based workflows. • Database caching layer: Caches data in your own database, partitioned per tenant and updated via webhooks and polling, so reads hit the database rather than the third-party API. • MCP support: Exposes an MCP layer on top of the REST API for agent tool use. 【Use cases】 1. Connect AI agents to Gmail, Slack, Notion, HubSpot, and other apps. 2. Send calendar invites and emails from an agent. 3. Pull records from Airtable or Sheets. 【Pricing】 Hobby: $0/month — For small projects, unlimited tool calls, 50 connections, 100k webhook events. Pro: $200/month — For teams in production, unlimited tool calls, unlimited connections, unlimited webhook events. Enterprise: Custom — For custom needs, unlimited tool calls, custom connections, custom webhook events. 【Pros and cons】 Pros: - Saves engineering time by removing the need to build per-provider auth and API infrastructure. - Credential isolation means agents never see API keys, which is a strong security posture for multi-tenant products. - The caching layer reduces latency and API rate-limit pressure by serving reads from your own database. - Open source and self-hostable, giving teams full control over their integration stack. Cons: - Custom integrations are community-contributed on the Hobby plan; Corsair team support is only on Pro, and custom integrations are available on Enterprise. - Avoiding any hosted dependency requires self-hosted execution. - The Hobby plan limits connections to 50 and webhook events to 100k. 【FAQ】 Q: Is Corsair free to use? A: Corsair offers a free Hobby plan with unlimited tool calls, 50 connections, and 100k webhook events. Paid plans start at $200/month for production teams. Q: Who is Corsair for? A: Corsair is designed for AI product teams, developers building AI agents, and teams in production that need multi-tenant integrations. Q: Does Corsair support self-hosting? A: Yes, you can run the full open source SDK on your own infrastructure at no cost. Self-hosted execution is required to avoid any hosted dependency.
Hobby · $0/month (For small projects, Unlimited tool calls, 50 connections) Pro · $200/month (For teams in production, Unlimited tool calls, Unlimited connections) Enterprise · Custom (For custom needs, Unlimited tool calls, Custom connections)
WPS AI is an AI assistant built into WPS Office that helps users write documents, generate presentation outlines and…
Baidu Comate is an AI coding assistant built by Baidu on its ERNIE model. It goes beyond simple autocomplete: it…
UniConverter is an AI media conversion and enhancement tool for video, audio, and images. It combines everyday…
6 similar tools to consider instead of Corsair