Search results

Benchmark: enforce headline eligibility for real-world episode claims

taskImplementation taskImplementation task

verificationRequires explicit verification evidenceRequires explicit verification evidence

Status: Open.

#1310 In shaun0927/openchrome;

· shaun0927 opened on May 16, 2026

Benchmark #D follow-up: inject reliability faults inside real-world tasks

reliabilityReliability and stability improvementReliability and stability improvement

taskImplementation taskImplementation task

Status: Open.

#1303 In shaun0927/openchrome;

· shaun0927 opened on May 15, 2026

Benchmark #D follow-up: real-world task completion as primary reliability signal

reliabilityReliability and stability improvementReliability and stability improvement

taskImplementation taskImplementation task

Status: Open.

#1304 In shaun0927/openchrome;

· shaun0927 opened on May 15, 2026

Benchmark #B follow-up: native/passive competitor adapter matrix

enhancementNew feature or requestNew feature or request

taskImplementation taskImplementation task

verificationRequires explicit verification evidenceRequires explicit verification evidence

Status: Open.

#1302 In shaun0927/openchrome;

· shaun0927 opened on May 15, 2026

Benchmark #B follow-up: real LLM repetitions and full-task metrics gate

enhancementNew feature or requestNew feature or request

taskImplementation taskImplementation task

verificationRequires explicit verification evidenceRequires explicit verification evidence

Status: Open.

#1301 In shaun0927/openchrome;

· shaun0927 opened on May 15, 2026

Benchmark #B follow-up: controlled realistic Agent Task Success workflow suite

enhancementNew feature or requestNew feature or request

taskImplementation taskImplementation task

verificationRequires explicit verification evidenceRequires explicit verification evidence

Status: Open.

#1300 In shaun0927/openchrome;

· shaun0927 opened on May 15, 2026

Benchmark: Episode-level token cost to completion

enhancementNew feature or requestNew feature or request

performancePerformance, latency, throughput, or resource-use improvementPerformance, latency, throughput, or resource-use improvement

taskImplementation taskImplementation task

Status: Open.

#1299 In shaun0927/openchrome;

· shaun0927 opened on May 15, 2026

Benchmark #F: Developer Experience — LOC/task, tool-schema quality, error actionability

enhancementNew feature or requestNew feature or request

taskImplementation taskImplementation task

Status: Open.

#1261 In shaun0927/openchrome;

· shaun0927 opened on May 14, 2026

Benchmark #E: Auth & Real-World Usability — logged-in success + setup cost

enhancementNew feature or requestNew feature or request

taskImplementation taskImplementation task

Status: Open.

#1260 In shaun0927/openchrome;

· shaun0927 opened on May 14, 2026

Benchmark #D: Reliability & Fault-Recovery — recovery rate, flaky rate, leak/zombie

enhancementNew feature or requestNew feature or request

reliabilityReliability and stability improvementReliability and stability improvement

taskImplementation taskImplementation task

Status: Open.

#1259 In shaun0927/openchrome;

· shaun0927 opened on May 14, 2026

Benchmark #C: Speed & Throughput — effective (success-weighted) throughput

enhancementNew feature or requestNew feature or request

performancePerformance, latency, throughput, or resource-use improvementPerformance, latency, throughput, or resource-use improvement

taskImplementation taskImplementation task

Status: Open.

#1258 In shaun0927/openchrome;

· shaun0927 opened on May 14, 2026

Benchmark #B: Agent Task Success — WebVoyager at equal LLM and equal budget

enhancementNew feature or requestNew feature or request

taskImplementation taskImplementation task

verificationRequires explicit verification evidenceRequires explicit verification evidence

Status: Open.

#1257 In shaun0927/openchrome;

· shaun0927 opened on May 14, 2026