The Prompting Company

Can an agent use your product?

Tell us which product to benchmark. The TPC team runs the benchmark on request: a coding agent reads the docs, runs a real use case in a fresh sandbox and an independent judge checks the result. The verdict and agent experience score are posted on the leaderboard.

Sorted by overall score: the average of experience and agent readiness. Completion says whether runs succeeded.

#ProductOverall
Loading benchmark results...
01ExaSearch100% passed
99
02context.devSearch100% passed
94
03SupabaseManaged database100% passed
90
04ResendEmail100% passed
90
05FirecrawlSearch100% passed
89
06NeonManaged database100% passed
88
07Browser UseBrowser Automation100% passed
87
08TursoManaged database100% passed
84
09PostmarkEmail100% passed · 1 not supported
84
10LoopsEmail100% passed · 2 not supported
82
11AgentMailEmail100% passed
81
12XataManaged database100% passed
81
13ParallelSearch100% passed
80
14PlanetScaleManaged database100% passed
78
15TavilySearch100% passed
76
16KernelBrowser Automation100% passed
75
17apollo.ioGTM Data100% passed
72
18BrowserbaseBrowser Automation100% passed
71
19SteelBrowser Automation100% passed
71
20Anchor BrowserBrowser Automation100% passed
70
26 products
Page 1 of 2