ruflo updates — week of 2026-05-31 (71 new commits, tip ef4acc49d)

[github-actions]

ruvnet/ruflo — last 7 days

Window: since 2026-05-24 · Upstream tip: ef4acc49d

feat (31)

• ef4acc49d feat(beir): 4-dataset BEIR (SciDocs joins) + config divergence finding (ADR-091, 3.10.30) (ruv)
• 05bb9cf7e feat(beir): 3-dataset BEIR + ruvector@0.2.27 + #2246 fixes (ADR-089/090, 3.10.29) (ruv)
• aa810856f feat(beir): Lucene BM25 + RRF + CE rerank — passes the acceptance test (ADR-088, 3.10.28) (ruv)
• ea99215f8 feat(beir): RRF ablation harness + HONEST NEGATIVE RESULT (ADR-087, 3.10.27) (ruv)
• 149027817 feat(intelligence): BEIR matrix + bootstrap CIs + 2nd dataset (ADR-086, 3.10.26) (ruv)
• 004cc2524 feat(intelligence): BEIR public benchmark + BGE bi-encoder — rank 2/11 on NFCorpus (ADR-085, 3.10.25) (ruv)
• 709b85949 feat(intelligence): cross-repo generalisation proof (ADR-084, 3.10.24) (ruv)
• d3223dc3d feat(intelligence): joint rerank re-grid + conditional defaults (ADR-083, 3.10.23) (ruv)
• fc39e29f2 feat(intelligence): grid-search retrieval defaults vs labelled metric (ADR-082, 3.10.22) (ruv)
• 17ce6ba67 feat(intelligence): labelled held-out corpus + nDCG metrics (ADR-081, 3.10.21) (ruv)
• 57a4a1b73 feat(intelligence): cross-encoder reranker (ADR-080, 3.10.20) (ruv)
• e6b557cb3 feat(intelligence): multi-field BM25 + opt-in type penalty (ADR-079, 3.10.19) (ruv)
• 62df8a184 feat(intelligence): hybrid retrieval + outcome signal (ADR-078, 3.10.18) (ruv)
• 0aac1e13f feat(intelligence): pretrain self-learning from repo history (ADR-077, 3.10.17) (ruv)
• 2b9e2de89 feat(intelligence): Round B (post-edit/-command + trajectory-end → globalStats) + Round C structured distillation (ADR-076) (ruv)
• ca77f8307 feat(intelligence): unified learning-stats aggregator (ADR-075, #2245 follow-up) (ruv)
• aca2280f1 feat(intelligence): self-learning wiring + honest multi-path output (ADR-074, #2245) (ruv)
• 0403a3427 feat(cost-tracker): wire codemod benchmark into cost-trend (ADR-143 follow-up) (ruv)
• 0988d92ce feat(intelligence): deterministic Tier-1 codemods replacing Agent Booster (ADR-143) (ruv)
• fd483f28d feat(workflows): add full-system-test native workflow (ruv)
• c983c0d80 feat(intelligence): Hermes-Agent Tier-1 adoptions — prompt caching, reasoning scrub, tool-loop breaker (3.10.10) (#2237) (rUv)
• 313f2f88e feat(intelligence): per-task bandit priors (ADR-142) + honest EWC/HNSW/MicroLoRA labeling (3.10.9) (#2231) (rUv)
• b8a49cfb8 feat(statusline): configurable cost symbol and hide toggle (#2220) (Wietze)
• 3ef6e175d feat(gaia): #2156 iter 63 — convergence layer (34/53, +2.5 vs baseline) (rUv)
• 663f5fe24 feat(gaia): #2156 iter 62 — convergence layer (deterministic finalization) (ruv)
• cea20282d feat(gaia): #2156 iter 53b — attachment tools (xlsx, pptx, py, png, mp3) Track T1 from Gate 1 (rUv)
• 427308308 feat: #2105 CLAUDE_FLOW_DB_PATH + #2058 team gateway checklist (#2146) (rUv)
• 19984c206 feat(ci): #2132 — cross-platform audit for plugin hooks.json commands (ruv)
• 31809ebeb feat(ci): #2127 — guard against stale alpha-tagged wrapper dep ranges (ruv)
• edde98f9e feat(graph): ADR-130 — unified graph intelligence backend (P1-P6) (#2129) (rUv)
• 47a7825b0 feat(rvagent): #ADR-129 — full rvagent integration (4 phases) (#2123) (rUv)

fix (17)

• 455152da0 fix(intelligence): 4-issue bug cluster — Opus 4.8 + uptime + MCP orphan + Q-encoder (ruv)
• 6198abfab fix(plugins): reconcile stale smoke contract assertions across 23 plugins (ruv)
• a73a2cbe3 fix(routing): stale route cache + --explore false (3.10.8) (#2229) (rUv)
• 08b362375 fix+perf+docs(intelligence): self-learning audit, hardening fixes, honest perf numbers (3.10.7) (#2228) (rUv)
• bb3489665 fix: #2219 #2221 #2222 #2215 #2226 — memory/routing/statusline bug cluster (3.10.6) (#2227) (rUv)
• 02aa8ed53 fix(init): #2206 #2207 #2208 — wizard server key, init detector, CLAUDE.md backup (#2214) (rUv)
• 2158808f7 fix(gaia): #2156 iter 53a — narrow T2 extraction to fix iter 52b regression (#2204) (rUv)
• a042ded01 fix(publish-workflow): build full v3 workspace before @claude-flow/cli (ruv)
• 2b9c0e714 fix(statusline): #2195 — generator delegates to hooks-statusline --json; fix ADR count, vectors, intelligence (#2196) (rUv)
• a6dd4ab3d fix(plugins/ruflo-core/hooks): #2155 — wrap 3 unwrapped .sh hooks in /bin/bash -c (#2161) (rUv)
• 733ada1b9 fix(release): #2151 — three-way version lockstep + CI guard (#2153) (rUv)
• 60f37f2d3 fix(ci): graph benchmark process.exit — eliminate 40-min hang (#2148) (rUv)
• 6778628e9 fix(security): #security-review v3.10 — bounds-check decodeEmbedding + env-var-precedence CI guard (#2144) (rUv)
• cdd5308d8 fix(hooks): #2132 — Windows-compatible Node shim + init-time platform detection (#2136) (rUv)
• 792044633 fix(ruflo): #2127 — tighten @claude-flow/cli dep from ^3.7.0-alpha.11 → ^3.10.0 (ruv)
• 16810c3e2 fix(bench): ADR-130 P6 — CI-friendly benchmark using single-session inserts (ruv)
• cfc341706 fix(memory): #2120 — accept NULL status in memory_entries (legacy DBs) + 3.7.0 stable (ruv)

docs (5)

• 951cc320f docs: ruflo-workflows v0.4.0 native-Workflow orchestration + ADR-141 (taubench campaign) (ruv)
• 6efe68fa1 docs(adr): ADR-129 status → Accepted + Gap 2 implementation record (#2201) (rUv)
• e7005d4d2 docs(gaia): submission reproducibility package (stable 34/53 config) (#2213) (rUv)
• 542481053 docs(adr): #ADR-130 — graph intelligence integration + improvement roadmap (ruv)
• 899163cb8 docs(readme): link Benchmarks gist + SOTA internals from Documentation section (ruv)

chore (13)

• d3771e1cf chore(release): bump 3.10.13 → 3.10.14 (ADR-074 self-learning wiring) (ruv)
• e6dc21fc7 chore(release): bump 3.10.12 → 3.10.13 + agentdb@3.0.0-alpha.16 (ruv)
• 367cb82ad chore: remove conflicting root overrides.agentdb (kept in ruflo wrapper) (ruv)
• cea01866e chore(release): bump 3.10.11 → 3.10.12 + agentdb@3.0.0-alpha.15 (closes #2235 A) (ruv)
• 48ca36948 chore(release): bump 3.10.10 → 3.10.11 (4-issue bug cluster) (ruv)
• 9085a4b34 chore: remove one-shot publish workflows (3.10.4 + 3.10.5 both published) (ruv)
• 9fa352ede chore(release): bump 3.10.4 → 3.10.5 (wizard init fixes #2206 #2207 #2208) (ruv)
• 38a016e1d chore(release): bump 3.10.3 → 3.10.4 (statusline fix #2195) + one-shot publish workflow (ruv)
• 3a919a60f chore(data): clone snapshot ruflo updates — week of 2026-05-17 (86 new commits, tip f8ab5a325) #4 — 0 clones (14d) (ruflo-bot)
• 3b0dc837a chore: project-wide quality sweep — no dead code, no slop, witness-validated (#2139) (rUv)
• 00f8a9fb2 chore(deps): bump pnpm/action-setup from 2 to 6 (dependabot[bot])
• e1bd1f072 chore: bump all packages to 3.10.0 — ADR-130 P4+P5+P6 (ruv)
• bf0f505c9 chore(release): v3.8.0 — ADR-129 rvagent full integration (ruv)

ci (1)

• 10086c4bb ci: add timeout-minutes: 40 to graph-benchmark job (ADR-130 P6) (ruv)

other (4)

• d29d87fb7 Merge fix/self-learning-2245: ADR-074 self-learning wiring (closes #2245) (ruv)
• 0545d7643 Merge feat/deterministic-tier1-codemods: deterministic Tier-1 codemods (ADR-143) (ruv)
• e7ec82f8c benchmarks: iter 56 — CodeAgent 53Q full measurement (30/53, 56.6%) (ruv)
• c9ec2b607 Merge pull request #2122 from ruvnet/fix/2120-memory-status-backfill (rUv)

---

Inspect a commit: git fetch upstream && git show <sha>
Diff a path: git diff upstream/main -- path/to/file
Port a commit: git cherry-pick <sha> (expect conflicts — no shared history)