state of agent readiness · this board

Four questions this board can answer

A dated cut of sites Waymark has scanned and kept. Not a census of the web. Numbers come from GET /api/research.

Latest stored scan: .

108
scanned
67.8
average
43.5%
B or above

Scanned board only — not a web census. Updated Sep 13, 2026.

01 / distribution

How scores sit on this board

Average 67.8, median 68.5. 7 A, 40 B, 56 C, 5 D. Bars are this board only — n=108, not a web census.

Score distribution

This board only, n=108 · median 68.5

n
108
Median
68.5
11
19
31
25
20
2
050100

Grades

Sites per band, this board

A
7 · 6.5%
B
40 · 37%
C
56 · 51.9%
D
5 · 4.6%

Category × grade

This board only, n=108

ai 20saas 17infra 11payments 7comms 6data 3
ABCDn
ai101020
saas8917
infra12811
payments167
comms2136
data123
other41718544

Protocol adoption

Share of this board with each file or handshake

llms.txt
78.7% · 85
OpenAPI
51.9% · 56
MCP
54.6% · 59

A map is a file. A contract is a spec. MCP is a live tool surface. Same n — this board only.

02 / who gets picked

Who gets picked

47 of 108 (43.5%) grade B or better. Those are the rows an agent can actually find a map or a tool surface on. n is 108 — we do not invent a web-wide census. A few named rows from the top of this cut:

Full board on /board.

03 / the line

What a B actually is

B means the official score cleared 70. Layers are Discovery 20, Access 30, Usability 40, Payments 10. A B is a working cut — maps plus enough door and tools — not proof an agent finished a job, and not a Deep Scan. A is 86. C is 48. On this board, 40 sites are B, 7 are A.

A+ ≥ 95 A ≥ 86 B ≥ 70 C ≥ 48 D ≥ 28 F below

Scan a site Full board How we score