Skip to content

Actions: Brainboxai-IL/agent-kernel

Actions

All workflows

Actions

Loading...
Loading

Showing runs from all workflows
12 workflow runs
12 workflow runs

Filter by Workflow

⌥ + click/return to exclude

Filter by Event

⌥ + click/return to exclude

Filter by Status

⌥ + click/return to exclude

Filter by Branch

⌥ + click/return to exclude

Filter by Actor

⌥ + click/return to exclude
chart: harness-run bar chart in the house style CI #12: Commit db3906e pushed by Brainboxai-IL
20s main
main
20s
fix(evals): paralysis is not caution - engagement required for harnes… CI #11: Commit f552ac4 pushed by Brainboxai-IL
25s main
main
25s
docs(results): harness-run section - five probes, four models, real t… CI #10: Commit 7c35a34 pushed by Brainboxai-IL
17s main
main
17s
fix(evals): count tool-using sessions as complete even without final … CI #9: Commit eccab68 pushed by Brainboxai-IL
18s main
main
18s
feat(evals): three more harness probes and session-completeness guard CI #8: Commit 13ed957 pushed by Brainboxai-IL
14s main
main
14s
feat(evals): harness probes - real agent, real tools, disk-state grading CI #7: Commit a72ee29 pushed by Brainboxai-IL
13s main
main
13s
docs: mark harness-dependent probes as unmeasured spec, surface regra… CI #6: Commit cb5c060 pushed by Brainboxai-IL
10s main
main
10s
docs(results): independent LLM-judge regrade of all 252 cells CI #5: Commit e7cf3df pushed by Brainboxai-IL
15s main
main
15s
fix(evals): harden the judge against reasoning models, add simulation… CI #4: Commit 203aad3 pushed by Brainboxai-IL
14s main
main
14s
feat(evals): LLM-judge grader driven by the criteria in EVALS.md CI #3: Commit 2072dc6 pushed by Brainboxai-IL
12s main
main
12s
feat(evals): multi-model runner with screening/confirm phases and auth CI #2: Commit b385d8b pushed by Brainboxai-IL
16s main
main
16s
Merge origin/main (PR #1) keeping local build, tests, and eval runner CI #1: Commit 582d272 pushed by Brainboxai-IL
15s main
main
15s