Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Agent Task Utility and Admission Integrity on Code-agent exact-effect benchmark
Loading...
231
STC Count
CXI
219.45
225.225
231
236.775
Jul 7, 2026
STC Count
Exact-Effect Authorization Success Rate
Manifest Verification Failure Rate
Capability Consumption Success Rate
Capability Lease Success Rate
Field Escapes Count
Effect Escapes Count
Invocation Escapes Count
Parse Failures Count
No Effect Authorization Count
Task Quality Failures Count
Updated 18d ago
Evaluation Results
Method
Method
Links
STC Count
Exact-Effect Authorization Success Rate
Manifest Verification Failure Rate
Capability Consumption Success Rate
Capability Lease Success Rate
Field Escapes Count
Effect Escapes Count
Invocation Escapes Count
Parse Failures Count
No Effect Authorization Count
Task Quality Failures Count
CXI
Unit / N=400 repo epis...
2026.07
231
-
-
-
-
0
0
0
73
61
35
Feedback
Search any
task
Search any
task