Is your documentation
ready for AI agents?
Agent Score evaluates how well your docs serve AI coding agents like Cursor, Claude and ChatGPT.
Developed by Fern Labs in partnership with Dachary Carey
Score your docs
Why it matters
Your docs have a new audience: AI is reading them now
Stripe 96·ElevenLabs 94·Square 89·Cloudflare 88·Anthropic 86·Vercel 85·Cohere 83·AWS 42·Slack 39·SAP 18·Stripe 96·ElevenLabs 94·Square 89·Cloudflare 88·Anthropic 86·Vercel 85·Cohere 83·AWS 42·Slack 39·SAP 18·
The invisible audience
AI coding agents are reading your API docs millions of times a day. If your docs aren't optimized for these agents, you're invisible to the fastest-growing segment of your user base.
Lighthouse for AI agents
Agent Score is the first industry benchmark for AI-agent readiness. Think Lighthouse, but for how effectively AI agents can discover, parse, and use your documentation. 0 to 100, 21 checks, 8 categories.
Agent readiness = competitive moat
API platforms with higher agent readability are already seeing outsized adoption. Agent readiness isn't a nice-to-have. It's the new SEO. 73% of top API docs still lack an llms.txt.
Open source, no black boxes
Built on the open-source Agent-Friendly Docs Spec with no gatekeeping, no black boxes. Every check is transparent and community-driven.
Agent score directory
How the top API documentation sites score on agent-readiness
Grade AAgent-ready
#1Resend86
A
11
Grade BGood
#2OpenAI77
B
21
#3Datadog73
B
14
Grade CImproving
#4Deepgram68
C
42
#5Cohere68
C
32
#6ElevenLabs68
C
42
#7docs-fiserv-dev-public-docs-guides64
C
51
#8buildwithfern-com-learn64
C
42
#9Auth064
C
15
#10Cloudflare64
C
24
Grade DNeeds work
#11LaunchDarkly59
D
44
#12Twilio59
D
53
#13Pinecone59
D
61
#14SendGrid55
D
63
#15Sentry55
D
4
#16Stripe55
D
45
#17Slack50
D
16
#18AWS50
D
17
#19Vercel50
D
63
#20platform-claude-com-docs50
D
46
#21shopify-dev50
D
63
Grade FNot ready
#22Shopify45
F
64
#23Clerk45
F
56
#24Plaid45
F
47
#25Supabase45
F
57
#26Anthropic45
F
56
#27MongoDB41
F
38
#28SAP32
F
6
#29Oracle Cloud32
F
24
#30Zendesk32
F
6
#31Modern Treasury32
F
6
#32Square32
F
48
#33HubSpot27
F
25
#34Salesforce27
F
25
21 checks across 8 categories
A comprehensive framework for evaluating agent-readiness
llms.txt
5 checks
Can agents discover and parse your documentation index?
Markdown availability
2 checks
Can agents get clean markdown instead of bloated HTML?
Page size
3 checks
Will your pages fit in an agent's context window?
Content structure
3 checks
Are tabs, headers, and code fences agent-parseable?
URL stability
2 checks
Do your URLs resolve cleanly without traps?
Agent discoverability
1 check
Can agents find your llms.txt from any page?
Observability
3 checks
Is your agent-facing content fresh and accurate?
Authentication
2 checks
Can agents access your docs without hitting auth walls?
Built by Fern
Fern-powered docs are agent-ready by default.
Trusted by
