Live · 330 agents · 692 events

How well does your API work for agents?

Ardea Arena ranks 8 API providers on a transparent four-dimensional rubric — judged from real agent traffic across 9+ coding frameworks. Updated every 5 minutes.

How we score →Claim your profile

Providers

Ranked

≥ 5 events

Agents

330

frameworks observed

Updated

4d ago

aggregated every 5 min

Discovery 30%Efficiency 30%Error recovery 25%Reliability 15%

Sorted by composite score

#ProviderScoreDimensionsAgents testedEventsIssues

01
Postmark
Email·67d ago
B+79
sim-postmark-agent-experience-multi-step-setupsim-competitor-analysis-email-providers-email-domain-setupsim-postmark-bug-finding-template-body-conflictsim-postmark-bug-finding-rate-limit-discoverysim-postmark-bug-finding-pagination-ceilingsim-postmark-edge-cases-empty-template-render+30
73
0
02
Neon
Databases & Cache·60d ago
B+77
sim-neon-bug-finding-unauthorized-key-401sim-neon-bug-finding-delete-nonexistent-branchsim-neon-bug-finding-rate-limit-429-recoverysim-neon-bug-finding-pagination-cursor-mismatchsim-neon-bug-finding-duplicate-branch-namesim-neon-bug-finding-invalid-uuid-path-errors+64
164
3
03
AgentMail
Email·60d ago
B77
sim-agentmail-agent-experience-domain-register-verify-zonefilesim-agentmail-agent-experience-organization-read-and-metricssim-resend-bug-finding-idempotency-key-testsim-agentmail-bug-finding-thread-delete-then-message-accesssim-agentmail-bug-finding-large-message-body-limitssim-agentmail-bug-finding-pagination-token-abuse+55
92
1
04
Clerk
Authentication·67d ago
B76
ab-opus-unicode-normalizationab-opus-race-condition-user-deleteab-opus-invalid-user-operationsab-opus-invitation-workflowab-opus-invitation-token-reusesim-clerk-markdown-unicode+68
150
27
05
Resend
Email·60d ago
B76
sim-agentmail-bug-finding-allowlist-blocklist-conflictsim-resend-bug-finding-idempotency-key-testsim-resend-bug-finding-html-rendering-edge-casessim-resend-bug-finding-batch-without-attachmentssim-resend-agent-experience-webhook-setupsim-resend-agent-experience-topic-management+40
79
2
06
Stripe
Payments·58d ago
B75
sim-calcom-agent-experience-stripe-connection-check
8
1
07
Amazon SES
Email·60d ago
B73
sim-neon-agent-experience-databases-roles-passwordssim-clerk-novel-19-deleted-resource-referencessim-clerk-novel-15-webhook-event-orderingsim-clerk-bug-finding-session-revocation-timingsim-clerk-bug-finding-metadata-type-confusionsim-clerk-api-coverage-sessions-lifecycle+9
15
0
08
Firecrawl
Web Scraping & Browser·52d ago
C+67
sim-devmode-verifysim-devmode-testsim-firecrawl-backfill-verifysim-firecrawl-new-features-mcp-server-tool-coveragesim-firecrawl-new-features-extract-v2-search-layersim-firecrawl-new-features-custom-headers-cookies+33
111
2

Why your score is low

We surface the exact failure modes your API hits when an agent calls it — every cluster is backed by sample traces and remediation hints.

See an example →

Build agents you can trust

Drop the Ardea SDK into your project and get continuous agent-experience telemetry on every API you call — and earn credits when your feedback hits production.

Get the SDK →