2026-08-06 - 2026-09-06
Overview
12 Pull requests merged by 3 users
Merged
#19 feat: add adaptive review effort budgets
Merged
#17 refactor: split opencode runtime and tests
Merged
#16 refactor: organize pilot modules and tests
Merged
#15 Refactor pilot architecture and remove dashboard
Merged
#14 test: eval judges e2e
Merged
#13 feat(eval): LLM-as-judge evaluators, dataset item fixes, and Experiments runs
Merged
#12 feat(pilot): behavioural scorers, feedback ground truth, and an eval dataset
Merged
#11 feat(pilot): emit per-review Langfuse traces
Merged
#10 fix(webhook): re-read PR labels at review start so a late AI-USAGE counts
Merged
#9 feat(input): add diff_compress module + prior-review compaction helpers
Merged
#7 harden the pilot against hostile PR content, add review skills + a cost model
Merged
#6 feat: token-usage reporting (AI-USAGE label)
1 Issue created by 1 user
Opened
#8 pragent feedback roll-up