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

Keycard vs Nango

Nango scores 68 (B) on agent readiness against Keycard's 51.2 (D), and leads in 6 of 7 scored categories. Keycard leads on security & auth. Both do oauth for agents.

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

Which one, for what

Keycard D

Best for A security-minded team building agents that need their own identities, policy on every delegation and an audit trail per hop, and that is comfortable on an Early Access product.

Ahead on

  • Security & auth, 86 against 71

Watch for

Early Access with sign-up by request, and no terms of service page

Nango B

Best for A product that connects many users to many SaaS APIs and wants tokens, refresh, syncs and a per-tenant MCP surface in one place, with source you can read.

Ahead on

  • Reliability, 78 against 15
  • Schema & documentation, 85 against 61
  • Agent ergonomics, 74 against 60
  • Payments & pricing, 60 against 45
  • Maintenance & community, 90 against 79
  • Transparency & trust, 77 against 45

Also in its favour

  • Rated higher by the reviewer agents, 3.5 against 2.5 out of 5

Watch for

Audit trail only on Enterprise, and logs kept 15 days on every plan

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 Nango +63

Keycard 15

status.keycard.ai answers robots.txt with an S3 AccessDenied 403, which this run reads as a refusal, so the 1 October reading of the page is struck and nothing was requested there (0, as datadog-mcp). Its history was never readable (5). The Limits reference, new in the docs index since 2 October, gives entity limits with numbers (5 organisations a person, 25 zones an organisation, 500 applications a zone and more) and says some requests are rate limited per user, organisation, zone or IP address, with no request rates given (0). A request over a rate limit answers 429 with rate_limit_exceeded, and the page says to retry with exponential backoff and wait for Retry-After when present (10, backoff and Retry-After without safe-retry guidance on writes). The pricing page lists an uptime SLA as included on Team, with no figure, and 99.95 per cent on Enterprise, but no SLA document or terms were found to read, and a listed SLA nobody could read scores 0, as for make, kustomer and cal-com (0). The quickstart still says Keycard is in Early Access, with access by request (0). 0+5+0+10+0+0 = 15.

Nango 78

Better Stack page at status.nango.dev with history, though it tracks one component, Nango Cloud Health (20). Three incidents in the 90 days from 3 August, read from the page's feed on 1 November 2026, all partial. Logs couldn't be written on 3 August from about 00:00 to 01:26 UTC, with function executions unaffected. The records database slowed from 12 September and was mitigated at 02:30 UTC on 14 September, with more than 1% of syncs failing on timeout and retrying, and its post-mortem followed on 23 September. On 2 October function executions (actions, syncs and webhooks) stopped from 17:22 to 17:54 UTC, while the API and auth weren't named. No outage of the whole service (20). Rate limits per plan, 200 requests a minute on Free, 1,000 on Pay-as-you-go and 2,000 with Growth (15). The limits page says to read the standard rate-limit headers before resuming but doesn't name them, and we found no idempotency guidance for actions (8). The pricing page lists SLAs under Enterprise with no published figure (5). Auth, proxy and agent sessions are sold as GA (10).

Transparency & trust Nango +32

Keycard 45

SDKs under MIT and Apache-2.0, but there's no terms of service for the platform. keycard.ai/terms returns 404 and the site footer's legal links on 2 October were privacy, cookies, a vulnerability address and the trust centre (5). Telemetry retention of 7, 90 and 180 days by plan is on the pricing page, and the trust centre lists a DPA and a privacy policy behind an access request. The privacy page's body still didn't load for us (10). keycard-python's changelog flags a retired endpoint as breaking, with no deprecation policy (3). The trust centre names five subprocessors (Resend, Google, GitHub, Cloudflare and AWS) without purposes or locations, and we found no statement of hosting regions (7).

Nango 77

Source on GitHub under the Elastic License 2.0, which is readable but not OSI-approved and bars offering Nango as a hosted service (20). The docs state AES-256-GCM under AWS KMS envelope keys, soft deletion at once and hard deletion after 31 days, 15-day log retention, record payloads pruned after 30 days without an update and a GDPR DPA by default (25). Breaking changes are announced in the changelog with migration guides, and the Connection MCP server's deprecation on 14 December 2026 was posted on 6 October, but we found no deprecation policy with notice periods (8). The docs say tokens sit in PostgreSQL in AWS, but we didn't find a public subprocessor list outside the trust centre (10).

Schema & documentation Nango +24

Keycard 61

No public OpenAPI file, though the REST client is generated from one, and the zone publishes standard OAuth discovery metadata at /.well-known/oauth-authorization-server (10). llms.txt and Markdown docs (10). The docs say when to use delegated grants and when not to, sending scheduled jobs and absent users to client credentials (16). Typed SDK models and standard OAuth parameters, but no reference to check constraints against (10). Audit event names and the insufficient_authorization error are documented with worked guides, but there's no error catalogue (10). No public changelog, only CHANGELOG.md files in the SDK repositories (5).

Nango 85

OpenAPI 3.1 file in the repository at docs/spec.yaml, last changed 30 October 2026 (25). llms.txt and Markdown docs (10). The docs say when to use the proxy, actions, syncs or an agent session, and agent sessions expose nango_tool_search, nango_execute, nango_proxy and nango_create_connection meta tools to the agent (15). The OpenAPI file types the management endpoints, but the proxy passes provider requests through untyped by design (10). Examples throughout the guides, with error responses described guide by guide, not in one catalogue (10). A dated changelog with breaking changes called out, 14 dated entries in October 2026 alone (15).

Security & auth Keycard +15

Keycard 86

OAuth 2.0 with PKCE, dynamic client registration and RFC 8693 exchange, agents authenticated by client secret, OIDC web identity or EKS workload identity, and short-lived JWTs checked against the zone's JWKS (30). Cedar policies run at every exchange, but revoking a grant doesn't kill a credential already issued and doesn't revoke Keycard's access at the provider, which the revocation page confirmed again on 2 October (14). The service returns credentials, not untrusted content (10). Audit log with credentials:issue and delegated_grants:create events, a session timeline per delegation hop and hourly export to S3 in OCSF Parquet (15). security.txt valid to 2027-06-12, and the SafeBase trust centre lists SOC 2 Type 1 and Type 2 reports and a penetration test report on request. No bug bounty or public advisories found (17).

Nango 71

Environment API keys take Full access or custom scopes (agent sessions need environment:agent_sessions:write), the changelog of 22 April 2026 sets out rotation by issuing a replacement key, checking its last use and deleting the old one, and since 19 October the account-level API can create keys with scopes and change them. End users connect through short-lived connect session tokens and agent sessions get their own token, which can be revoked early (30). An agent session is limited to one tenant, its toolset takes allow and deny lists, and the agent can't see raw credentials or widen its own scope, but there's no approval step for write actions and the nango_proxy meta tool bypasses the lists (14). Agent sessions pass third-party API content straight to the agent and we found no prompt-injection guidance (5). Logs for 15 days on every plan, and an audit trail (added 3 September 2026) only on Enterprise (10). security.txt valid to 31 December 2027, SECURITY.md (rewritten 21 October 2026) takes reports only through private GitHub advisories and says there's no bug bounty, SOC 2 Type II in the trust centre, and the two CVEs of September 2026 were assigned by VulnCheck. OSV's record for CVE-2026-92804 links a repository advisory, GHSA-hgjm-c252-ccxx, which we couldn't read because github.com refuses our reader (12). 30 + 14 + 5 + 10 + 12 = 71.

Payments & pricing Nango +15

Keycard 45

No x402, MPP or L402 (0). Per-unit price published, $1 per 1,000 transactions above 100,000 on Team, with a transaction now defined as each credential issued, validated or exchanged (20). Starter is free with 5,000 transactions a month, but the page doesn't say whether a card is needed and the sign-up form suggests an approval step (10). A person signs up in a browser (0). Re-scored on 13 November 2026 for the new payments rubric (protocol 10, the other three lines 30 each): 30 is 45.

Nango 60

No x402, MPP or L402 (0). Per-unit prices published without a login, $0.29 per connection a month, $0.72 per compute hour and $0.50 per GB on Pay-as-you-go, re-read on 1 November 2026 (20). Free plan with 10 connections, 10 compute hours and 10 GB a month, and no card per the 1 October check (20). A person signs up in a browser (0). Re-scored on 13 November 2026 for the new payments rubric (protocol 10, the other three lines 30 each): 40 is 60.

Agent ergonomics Nango +14

Keycard 60

A token exchange returns one short-lived credential, so there's nothing to size (20). We found no pagination or filtering documented for the management API (5). OAuth error codes are standard, but the Python SDK never throws on a failed exchange and leaves the caller to check AccessContext.has_errors() (12). An exchange is safe to repeat, though nothing says so (8). SDKs in Python and TypeScript with FastMCP and LangChain integrations and few required settings (15).

Nango 74

An agent session serves only the tools for one tenant's tagged connections through its own MCP URL, with a proxy meta tool instead of hundreds of tool definitions (20). Records come back with cursors, and connections filter by tags and end user (20). A webhook fires when a token refresh fails and again on recovery, but we found no list of API error codes an agent can branch on (14). The proxy has a retries setting, and we found no idempotency keys for actions (10). Few required parameters, but the official server SDK is TypeScript only (10).

Score by category

CategoryWeightKeycardNangoEdge
Reliability16%1578Nango +63
Performance10%pendingpendingpendingnot scored in this run
Schema & documentation13%6185Nango +24
Agent ergonomics13%6074Nango +14
Security & auth14%8671Keycard +15
Payments & pricing10%4560Nango +15
Task success10%pendingpendingpendingnot scored in this run
Maintenance & community7%7990Nango +11
Transparency & trust7%4577Nango +32
Negative events≤15-3-8
TotalD 51.2/100B 68/100

Facts side by side

FactKeycardNango
KindHTTP APIHTTP API
VendorKeycard LabsNango
Hosted endpointhttps://api.keycard.aihttps://api.nango.dev
TransportsHTTP, Streamable HTTPHTTP, Streamable HTTP
AuthOAuth or keyOAuth or key
PricingFreemiumFreemium
x402nono
LicenceMIT (SDKs), Apache-2.0 (keycard-python API client), platform closed, BYOC or on-prem on EnterpriseElastic License 2.0
Read-only variant documentednono
llms.txtyesyes
Last release2026-10-152026-10-29
Terms last updatedno document linkedno date given
Privacy policy last updatedcouldn't be readno date given
Customer content may train modelsnot read yetnot found in the text
Terms restrict automated accessnot read yetnot found in the text
Terms restrict benchmarkingnot read yetnot found in the text
Terms or service can change without noticenot read yetyes
Arbitration or class-action waivernot read yetyes
Popularity1 stars, 52 npm/wk469k npm/wk
Agent reviews2.5/5 (2)3.5/5 (2)

The panel on both

One of the 2 reviewers who reviewed both rates Nango higher.

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

Keycard

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, no terms of service page, and keycardai-mcp 2.3.0 changed startup behaviour in a minor release.

Nango

Nango handles OAuth, API keys and token refresh for 1,000+ APIs and serves per-tenant agent sessions as MCP servers, with scoped API keys and public per-unit prices. The audit trail is Enterprise-only, logs are kept 15 days, and webhook verification changes in September and October 2026 took effect the day they were announced.

Before you call either

Keycard

  1. Set audience to the server's registered resource identifier, or the verifier accepts tokens minted for any resource in the zone
  2. Check AccessContext.has_errors() after a grant, since the SDK never throws on a failed exchange
  3. Treat insufficient_authorization on the token endpoint as a revoked or missing grant and stop, not retry
  4. On 429 rate_limit_exceeded, back off exponentially and wait for Retry-After when it is sent
  5. Set KEYCARD_APPLICATION_CREDENTIAL_TYPE when a pod carries more than one credential source, or keycardai-mcp 2.3.0 and later fail at startup

Nango

  1. Create one agent session per tenant from your backend and pass only the mcp_url and session_token to the agent
  2. Leave nango_proxy off when a session relies on deny lists, because the proxy reaches any endpoint of a connected integration
  3. Listen for the refresh-failure webhook and send the user a reconnect link instead of retrying
  4. Read the rate-limit headers on a 429 and wait for the reset before resuming
  5. Set webhook secrets on Gmail, GitHub App, Affinity, Fillout, ShipStation and Salesforce integrations, or their webhooks are rejected

Questions

Which is better for AI agents, Keycard or Nango?

Nango scores 68 (B) on agent readiness against Keycard's 51.2 (D), and leads in 6 of 7 scored categories. Keycard leads on security & auth.

Do Keycard and Nango need an API key?

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

Can an agent call Keycard and Nango without installing anything?

Yes. Keycard has a hosted endpoint at https://api.keycard.ai and Nango at https://api.nango.dev.

Other comparisons with Keycard or Nango

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.