2026-01-02 - 2026-07-02
Overview
1 Release published by 1 user
Published
v1
19 Pull requests merged by 1 user
Merged
#20 feat: re-platform agentic review onto executus + large-PR cost controls
Merged
#18 feat: inline COMMENT-state PR review (findings anchored to changed lines)
Merged
#17 feat: cross-model consensus consolidation (one ranked comment, not N walls)
Merged
#16 feat: structured findings contract (machine-readable gadfly-findings block)
Merged
#15 docs: correct examples/reusable.yml pin guidance (prefer @sha; runners cache @v1)
Merged
#14 feat(reusable): runtime-variable swarm config (cache-immune, no more re-pinning to retune)
Merged
#13 feat(reusable): add the 4090 Ti (qwen3.6-27b via llama-swap) to the default swarm
Merged
#12 docs: recommend the @v1 release tag for reusable-workflow consumers
Merged
#11 tune(reusable): claude-code=3 models × 5 lenses (claude was the bottleneck)
Merged
#10 feat(reusable): ship the curated swarm as the default config consumers inherit
Merged
#9 security: scope reusable-workflow secrets (least privilege) over secrets: inherit
Merged
#8 feat: Phase 4 — reusable "subscribe" workflow (+ dogfood it)
Merged
#7 feat: bump majordomo + support llama-swap(s) provider spellings
Merged
#6 ci: enable claude-code/opus:max (max-thinking) reviewer
Merged
#5 feat: add claude-code/opus reviewer + max-thinking spec support
Merged
#4 fix: clean-lens findings + trim the dogfood swarm to strong reviewers
Merged
#3 feat: claude-code backends + llamaswap provider + dogfood the CC engine
Merged
#2 feat: claude-code reviewer engine (per-lens claude -p shell-out)
Merged
#1 feat: live status-board comment — per-model/per-lens review progress