# Auth and delegated access for AI agents > 13 agent auth & delegated access ranked by the Anchor benchmark. Leader Descope Agentic Identity Hub (A). Services that let an agent act for a user in other apps: OAuth flows and token storage for third-party APIs, scoped and revocable access, and an identity for the agent itself. Compared on the providers they cover, how consent works, token handling and audit logs. - Canonical: https://www.anchorterminal.com/categories/agent-auth - Markdown: https://www.anchorterminal.com/categories/agent-auth.md (~4,550 tokens) - Slim: https://www.anchorterminal.com/categories/agent-auth.min.md (~680 tokens, same facts, less prose, for token-sensitive contexts) - JSON: https://www.anchorterminal.com/categories/agent-auth.json (this page as data, same URL with Accept: application/json) - Site index for agents: https://www.anchorterminal.com/llms.txt (full text: https://www.anchorterminal.com/llms-full.txt) - API: https://www.anchorterminal.com/api/v1/index.json - Updated: 2026-10-04 Services that let an agent act for a user in other apps: OAuth flows and token storage for third-party APIs, scoped and revocable access, and an identity for the agent itself. Compared on the providers they cover, how consent works, token handling and audit logs. - Tools ranked: 13 · agent-ready (BB or better): 4 · accept x402: 0 · hosted endpoints: 13 · desk reviews by the panel: 38 - JSON: https://www.anchorterminal.com/api/v1/tools.json (list) · https://www.anchorterminal.com/api/v1/rankings.json (ranked) · https://www.anchorterminal.com/api/v1/x402.json (payable) · https://www.anchorterminal.com/api/v1/capabilities.json (by capability) - Grades run AA, A, BB, B, C, D, E, F · methodology: https://www.anchorterminal.com/benchmark/ - Capabilities in this category: auth.oauth, auth.tokens, auth.consent, auth.agent-identity, auth.audit - https://letme.dev/auth.oauth picks the top-graded tool in this list and says how to call it direct; calling through letme comes later (https://www.anchorterminal.com/letme/index.md) ## Ranking | # | Tool | Vendor | Kind | Category | Grade | Score | Confidence | x402 | Auth | Where | Reviews | Page | | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | | 10 | Descope Agentic Identity Hub | Descope | HTTP API | Agent auth | A | 79.2 | medium | no | OAuth or key | hosted | 3.1/5 (8) | https://www.anchorterminal.com/tools/descope-agentic-identity.md | | 36 | Composio (API + MCP) | Composio | HTTP API | Tool access | BB | 75.3 | medium | no | OAuth or key | hosted | 3.5/5 (8) | https://www.anchorterminal.com/tools/composio-rube.md | | 74 | Scalekit AgentKit | Scalekit | HTTP API | Agent auth | BB | 72.1 | medium | no | OAuth or key | hosted | 2.5/5 (2) | https://www.anchorterminal.com/tools/scalekit-agentkit.md | | 82 | Auth0 for AI Agents (Token Vault) | Auth0 by Okta | HTTP API | Agent auth | BB | 71.5 | medium | no | OAuth | hosted + local | 3.5/5 (2) | https://www.anchorterminal.com/tools/auth0-ai-agents.md | | 135 | Nango | Nango | HTTP API | Agent auth | B | 67.9 | medium | no | OAuth or key | hosted | 3.5/5 (2) | https://www.anchorterminal.com/tools/nango.md | | 147 | Arcade.dev | Arcade.dev | HTTP API | Agent auth | B | 67 | medium | no | OAuth or key | hosted + local | 2.5/5 (2) | https://www.anchorterminal.com/tools/arcade.md | | 167 | Pipedream API + MCP | Pipedream (Workday) | HTTP API | Workflows | B | 65.8 | medium | no | OAuth or key | hosted | 2.5/5 (2) | https://www.anchorterminal.com/tools/pipedream.md | | 241 | Stytch Connected Apps | Stytch (Twilio) | HTTP API | Agent auth | C | 60.8 | medium | no | OAuth or key | hosted | 3/5 (2) | https://www.anchorterminal.com/tools/stytch-connected-apps.md | | 256 | WorkOS Pipes and Agents | WorkOS | HTTP API | Agent auth | C | 60 | medium | no | OAuth or key | hosted | 2.5/5 (2) | https://www.anchorterminal.com/tools/workos-pipes.md | | 280 | Zapier MCP (agent actions) | Zapier | MCP server | Tool access | C | 58.4 | medium | no | OAuth or key | hosted | 2.5/5 (2) | https://www.anchorterminal.com/tools/zapier-mcp.md | | 303 | Keycard | Keycard Labs | HTTP API | Agent auth | C | 56.3 | medium | no | OAuth or key | hosted | 2.5/5 (2) | https://www.anchorterminal.com/tools/keycard.md | | 321 | Permit MCP Gateway | Permit.io | Model platform | Human approval | C | 54.5 | medium | no | OAuth | hosted | 2.5/5 (2) | https://www.anchorterminal.com/tools/permit-mcp-gateway.md | | 381 | Paragon ActionKit + MCP | Paragon | HTTP API | Workflows | D | 47.8 | medium | no | OAuth or key | hosted | 2/5 (2) | https://www.anchorterminal.com/tools/paragon.md | Scores are from public evidence against the published checklist (https://www.anchorterminal.com/benchmark/), with Performance and Task success pending. p95 latency and context cost come from our probes, which haven't run yet. ## Summaries ### 10. Descope Agentic Identity Hub, A (79.2) Descope's identity and access tools for AI agents, built on its customer identity platform. Token vault for user and tenant tokens with scoped fetch, forced refresh and per-token deletion. No tool catalogue, so you write every provider call yourself. - Page: https://www.anchorterminal.com/tools/descope-agentic-identity · Markdown: https://www.anchorterminal.com/tools/descope-agentic-identity.md · JSON: https://www.anchorterminal.com/api/v1/tools/descope-agentic-identity.json - Capabilities: auth.oauth, auth.tokens, auth.consent, auth.agent-identity, auth.audit, hitl.approve · endpoint: `https://api.descope.com` ### 36. Composio (API + MCP), BB (75.3) Tool access and per-user authentication for agents across 1,000+ apps. Connect supports toolkit selection and filtering by read-only or destructive actions. Rube closed on 16 May 2026, so existing rube.app configurations need migration. - Page: https://www.anchorterminal.com/tools/composio-rube · Markdown: https://www.anchorterminal.com/tools/composio-rube.md · JSON: https://www.anchorterminal.com/api/v1/tools/composio-rube.json - Capabilities: automation.apps, automation.auth, automation.actions, agent.tools, automation.webhooks · endpoint: `https://backend.composio.dev/api/v3.1` ### 74. Scalekit AgentKit, BB (72.1) Authentication and integration platform for agents, with per-user account connections, scoped MCP servers and managed tool calls. 500+ connectors, including remote MCP servers over OAuth 2.1 with DCR. No rate limits or idempotency documented for Scalekit's own API. - Page: https://www.anchorterminal.com/tools/scalekit-agentkit · Markdown: https://www.anchorterminal.com/tools/scalekit-agentkit.md · JSON: https://www.anchorterminal.com/api/v1/tools/scalekit-agentkit.json - Capabilities: auth.oauth, auth.tokens, auth.consent, auth.agent-identity, agent.tools · endpoint: `https://{env}.scalekit.com` ### 82. Auth0 for AI Agents (Token Vault), BB (71.5) Auth0's identity and authorisation tools for AI agents, built on its identity platform. Standard grants throughout, RFC 8693 token exchange, CIBA with RAR and DPoP. Only works when Auth0 is the identity provider for your users. - Page: https://www.anchorterminal.com/tools/auth0-ai-agents · Markdown: https://www.anchorterminal.com/tools/auth0-ai-agents.md · JSON: https://www.anchorterminal.com/api/v1/tools/auth0-ai-agents.json - Capabilities: auth.oauth, auth.tokens, auth.consent, auth.agent-identity, hitl.approve · endpoint: `https://{tenant}.auth0.com/oauth/token` ### 135. Nango, B (67.9) Source-available integration platform that handles OAuth, API keys and token refresh for 1,000+ APIs on behalf of your users. 1,000+ APIs with OAuth, API key and client-credentials auth handled. Audit trail only on Enterprise, and logs kept 15 days on every plan. - Page: https://www.anchorterminal.com/tools/nango · Markdown: https://www.anchorterminal.com/tools/nango.md · JSON: https://www.anchorterminal.com/api/v1/tools/nango.json - Capabilities: auth.oauth, auth.tokens, auth.consent, auth.audit, agent.tools, automation.embedded · endpoint: `https://api.nango.dev` ### 147. Arcade.dev, B (67) MCP runtime built around per-user authorisation. 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. - Page: https://www.anchorterminal.com/tools/arcade · Markdown: https://www.anchorterminal.com/tools/arcade.md · JSON: https://www.anchorterminal.com/api/v1/tools/arcade.json - Capabilities: auth.oauth, auth.tokens, auth.consent, auth.audit, agent.tools · endpoint: `https://api.arcade.dev` ### 167. Pipedream API + MCP, B (65.8) Code-first workflows in Node.js, Python, Go and Bash, plus Connect, an API and SDK that runs 10,000+ prebuilt actions across 3,000+ apps on behalf of your own users with managed OAuth. Managed OAuth per end user across 3,000+ apps, through API, SDK or MCP. Public changelog's last entry is 1 October 2025. - Page: https://www.anchorterminal.com/tools/pipedream · Markdown: https://www.anchorterminal.com/tools/pipedream.md · JSON: https://www.anchorterminal.com/api/v1/tools/pipedream.json - Capabilities: automation.workflows, automation.apps, automation.embedded, automation.code, automation.webhooks, automation.auth, agent.tools · endpoint: `https://api.pipedream.com/v1` ### 241. Stytch Connected Apps, C (60.8) Turns a Stytch project into an OAuth 2.1 and OIDC authorisation server so agents and MCP clients can act for your users. OAuth 2.1 authorisation server with DCR, CIMD and PKCE out of the box. No outbound token vault, so it can't hold your users' third-party tokens. - Page: https://www.anchorterminal.com/tools/stytch-connected-apps · Markdown: https://www.anchorterminal.com/tools/stytch-connected-apps.md · JSON: https://www.anchorterminal.com/api/v1/tools/stytch-connected-apps.json - Capabilities: auth.oauth, auth.consent, auth.agent-identity, auth.tokens · endpoint: `https://api.stytch.com` ### 256. WorkOS Pipes and Agents, C (60) WorkOS tools for connecting agents to third-party accounts, managing access tokens and assigning revocable agent identities. Agent identity with per-session revocation and token lifetimes set per blueprint. 21 incidents on the status page since 3 July 2026, several over an hour. - Page: https://www.anchorterminal.com/tools/workos-pipes · Markdown: https://www.anchorterminal.com/tools/workos-pipes.md · JSON: https://www.anchorterminal.com/api/v1/tools/workos-pipes.json - Capabilities: auth.oauth, auth.tokens, auth.consent, auth.agent-identity, auth.audit · endpoint: `https://api.workos.com` ### 280. Zapier MCP (agent actions), C (58.4) Hosted MCP server that lets agents discover and run actions across apps connected to the user's Zapier account. 16 meta-tools in agentic mode, or managed mode with only the actions you pick. Connection tokens are long-lived and the docs allow them in the URL query string. - Page: https://www.anchorterminal.com/tools/zapier-mcp · Markdown: https://www.anchorterminal.com/tools/zapier-mcp.md · JSON: https://www.anchorterminal.com/api/v1/tools/zapier-mcp.json - Capabilities: automation.apps, automation.auth, automation.actions, agent.tools · endpoint: `https://mcp.zapier.com/api/v1/connect` ### 303. Keycard, C (56.3) Identity and access platform for AI agents. Agent identity by client secret, OIDC web identity or EKS workload identity, with Cedar policy at every token exchange. Early Access with sign-up by request, and no terms of service page. - Page: https://www.anchorterminal.com/tools/keycard · Markdown: https://www.anchorterminal.com/tools/keycard.md · JSON: https://www.anchorterminal.com/api/v1/tools/keycard.json - Capabilities: auth.oauth, auth.tokens, auth.consent, auth.agent-identity, auth.audit · endpoint: `https://api.keycard.ai` ### 321. Permit MCP Gateway, C (54.5) Hosted proxy between MCP clients and MCP servers that signs in the human behind the agent, checks each tool call against Permit.io policy and logs it. No SDK or client change, since the client points at the gateway URL and keeps its tool list. Approvals are Enterprise only, through a demo, with no published price. - Page: https://www.anchorterminal.com/tools/permit-mcp-gateway · Markdown: https://www.anchorterminal.com/tools/permit-mcp-gateway.md · JSON: https://www.anchorterminal.com/api/v1/tools/permit-mcp-gateway.json - Capabilities: hitl.approve, hitl.channels, hitl.audit, auth.oauth, auth.consent, auth.agent-identity, auth.audit · endpoint: `https://{subdomain}.agent.security/mcp` ### 381. Paragon ActionKit + MCP, D (47.8) Embedded integration platform for SaaS products. Per-end-user RS256 JWT on every call, with a hosted Connect Portal for OAuth. No published prices and no self-serve paid plan. - Page: https://www.anchorterminal.com/tools/paragon · Markdown: https://www.anchorterminal.com/tools/paragon.md · JSON: https://www.anchorterminal.com/api/v1/tools/paragon.json - Capabilities: automation.embedded, automation.workflows, automation.apps, automation.auth, automation.webhooks, agent.tools · endpoint: `https://actionkit.useparagon.com` ## How we test this category An agent connects to two third-party apps for a test user, makes calls, has one scope refused and then the grant revoked. We check the consent flow, where tokens live, what the audit log shows and how revocation reaches the agent. This test hasn't run yet, so Task success is pending and the grades here come from the categories assessed from public evidence. ## Indexed, not reviewed (23) Sorted into this category from public catalogues, with facts and our own checks but no score, grade or rank (https://www.anchorterminal.com/indexed/index.md). | Listing | Kind | What it does | Why it's here | | --- | --- | --- | --- | | [artifacta.io MCP server](https://www.anchorterminal.com/tools/artifacta-mcp.md) | MCP server | Artifact store for AI agents. Hosted OAuth at mcp.artifacta.io/mcp; local stdio via npm/PyPI. | vendor's own | | [Coach Watts](https://www.anchorterminal.com/tools/coachwatts-coach-watts.md) | MCP server | Remote MCP server for training, nutrition, wellness, and performance data with OAuth 2.0. | vendor's own | | [Expense Budget Tracker](https://www.anchorterminal.com/tools/expense-budget-tracker.md) | MCP server | Track expenses, budgets, balances, transfers, and multi-currency reports with OAuth-secured tools. | vendor's own | | [Gemina](https://www.anchorterminal.com/tools/gemina.md) | MCP server | Extract, search and tag any document: invoices, receipts, contracts, templates. OAuth or API key. | vendor's own | | [HitPay MCP Server](https://www.anchorterminal.com/tools/hit-pay-mcp.md) | MCP server | Official HitPay MCP: sales, payouts, balances; create payment links and invoices. OAuth; no refunds. | vendor's own | | [imaginevid-ai-generation](https://www.anchorterminal.com/tools/imaginevid-ai-generation.md) | MCP server | OAuth-protected ImagineVid MCP for image, video, and music generation. | vendor's own | | [kernel-mcp-server](https://www.anchorterminal.com/tools/onkernel-kernel-mcp-server.md) | MCP server | Access Kernel's cloud-based browsers and app actions via MCP (remote HTTP + OAuth). | vendor's own | | [Labby](https://www.anchorterminal.com/tools/dinglebear-labby.md) | MCP server | Rust MCP gateway with Code Mode, authentication, setup, logs, CLI, HTTP API, and operator web UI. | vendor's own | | [logi](https://www.anchorterminal.com/tools/1pass-logi.md) | MCP server | logi (1pass) IdP — manage your OAuth apps, redirect URIs, passkeys, login history and docs. | vendor's own | | [MCP Protocol Conformance](https://www.anchorterminal.com/tools/studiomeyer-protocol-conformance.md) | MCP server | MCP conformance test harness. JSON-RPC, OAuth 2.1 PKCE, schemas, smoke, annotations. CLI + lib. | vendor's own | | [mymlh-mcp-server](https://www.anchorterminal.com/tools/git-mymlh-mcp-server.md) | MCP server | OAuth-enabled MyMLH MCP server for accessing MyMLH data. | vendor's own | | [odoo](https://www.anchorterminal.com/tools/odooconsole-odoo.md) | MCP server | Odoo ERP for AI agents: hosted OAuth endpoint, gated writes, one endpoint for every instance. | vendor's own | | [OfflineCreator Studio](https://www.anchorterminal.com/tools/offlinecreatorstudio-mcp.md) | MCP server | Create images and videos with OfflineCreator Studio through OAuth or the npm stdio server. | vendor's own | | [OpenWork MCP Gateway](https://www.anchorterminal.com/tools/openworklabs-openwork.md) | MCP server | Your OpenWork org's skills, plugins, workflows, and connections through one OAuth MCP URL. | vendor's own, widely used | | [pAIchart MCP Hub](https://www.anchorterminal.com/tools/paichart-mcp-hub.md) | MCP server | MCP Hub: AI service discovery, per-user OAuth, and multi-service workflow orchestration | vendor's own | | [platform7n](https://www.anchorterminal.com/tools/p7n-platform7n.md) | MCP server | Connect Claude to your Platform7n workspaces — chat, links, and tasks. One-click OAuth. | vendor's own | | [Prizmad](https://www.anchorterminal.com/tools/prizmad-mcp-server.md) | MCP server | Generate AI UGC video ads from any product URL — avatars, voiceover, OAuth Connect. | vendor's own | | [thoughtspot.app MCP server](https://www.anchorterminal.com/tools/thoughtspot-mcp-server.md) | MCP server | MCP Server for ThoughtSpot - provides OAuth authentication and tools for querying data | vendor's own | | [toll402-mcp](https://www.anchorterminal.com/tools/toll402-mcp.md) | MCP server | Pay-per-call tools for agents: 2,600+ APIs, OAuth connector; pay with x402 or card credits. | vendor's own | | [WHOOP — MissingMCP](https://www.anchorterminal.com/tools/missingmcp-whoop.md) | MCP server | WHOOP recovery, strain, sleep and workouts in Claude via official WHOOP OAuth. Free, open source. | vendor's own | | [Wraps](https://www.anchorterminal.com/tools/wraps-docs.md) | MCP server | Search the Wraps docs and estimate AWS SES costs. Public, read-only, no authentication. | vendor's own | | [Xquik MCP Server](https://www.anchorterminal.com/tools/xquik-mcp.md) | MCP server | 128 REST operations. 120 MCP routes; 119 JSON/text ops. OAuth 2.1. Not affiliated with X Corp. | vendor's own | | [ynab-mcp-server](https://www.anchorterminal.com/tools/smirnovlabs-ynab-mcp-server.md) | MCP server | Hosted remote MCP server for YNAB on Cloudflare Workers with OAuth | vendor's own |