-
Notifications
You must be signed in to change notification settings - Fork 65
Pull requests: benchflow-ai/benchflow
Author
Label
Projects
Milestones
Reviews
Assignee
Sort
Pull requests list
fix(review): record escaping workspace symlinks as evidence exclusions
#1130
opened Sep 15, 2026 by
bingran-you
Collaborator
Loading…
fix(sandbox): upload denylist compose overlay on Daytona DinD
#1123
opened Sep 14, 2026 by
bingran-you
Collaborator
Loading…
fix(metrics): make collect_metrics selection predicate shape-aware fo…
#1121
opened Sep 12, 2026 by
VaisakhiMishra
Contributor
Loading…
fix(sandbox): reject pytest plugins backed by agent-writable code
#1117
opened Sep 9, 2026 by
tulerfeng
Loading…
URL blocklist network mode enforced inside the sandbox (Daytona, Docker)
#1115
opened Sep 9, 2026 by
zhiheng-yang
Loading…
feat(sandbox): enforce agent egress blocklist with in-container proxy and firewall
#1114
opened Sep 9, 2026 by
questiondlmarks
Loading…
feat: add private research controls on top of URL denylist
#1112
opened Sep 8, 2026 by
ElegantLin
Contributor
Loading…
fix: preserve safe ACP tool provenance in trajectories
#1111
opened Sep 8, 2026 by
kywch
Collaborator
Loading…
feat: expose custom rollout planes through SDK.run
#1110
opened Sep 8, 2026 by
kywch
Collaborator
Loading…
fix: disconnect agents before verification and harden cleanup
#1109
opened Sep 8, 2026 by
kywch
Collaborator
Loading…
Post-hoc cost from recorded usage + OAuth-mode usage capture (bench cost, pricing table, OTEL receiver)
#1108
opened Sep 7, 2026 by
yuchenlwu
Contributor
Loading…
Preserve Gemini passthrough trajectories in training exports
#1106
opened Sep 5, 2026 by
kywch
Collaborator
Loading…
Fix Gemini CLI Vertex routing and authentication
#1105
opened Sep 5, 2026 by
kywch
Collaborator
Loading…
fix(metrics): guard reward reads against a non-dict rewards field
#1096
opened Sep 4, 2026 by
Galius5136
Loading…
fix(rollout): route verifier timeout cleanup by service
#1095
opened Sep 4, 2026 by
zhiheng-yang
Loading…
fix(rollout): clear stale LLM trajectory on rollout reuse
#1094
opened Sep 3, 2026 by
zhiheng-yang
Loading…
refactor(agents): extract OpenClaw from BenchFlow core
#1093
opened Sep 3, 2026 by
kywch
Collaborator
Loading…
feat(review): integrate task rubrics into rollout scoring
#1092
opened Sep 3, 2026 by
bingran-you
Collaborator
Loading…
fix(sandbox): verify the sandbox-user kill before scoring
#1088
opened Sep 2, 2026 by
tulerfeng
Loading…
feat(continue): typed continue-gate errors and batch skips instead of silent drops
#1084
opened Aug 31, 2026 by
JeremyJC67
Loading…
fix(sandbox): keep timeout teardown from erasing the timeout
#1081
opened Aug 31, 2026 by
tulerfeng
Loading…
feat(sandbox): add Apptainer backend for HPC environments
#1073
opened Aug 31, 2026 by
zhiheng-yang
Loading…
Capture LLM trajectories across OAuth and API-key runs
#1057
opened Aug 29, 2026 by
bingran-you
Collaborator
Loading…
Previous Next
ProTip!
Mix and match filters to narrow down what you’re looking for.