Every commit scored for change-risk against this repo's own history, so 'elevated' means elevated here rather than on some global curve.
Needs review
42 commits sit in this repo's top risk tercile, which is 30% of the 138scored. The cut is drawn against this codebase's own history rather than a global curve, so a quiet repo still fills its top band, and here it starts at 9.6 out of 10. What pushes a commit up is size and spread together: a large change confined to one area scores below a smaller one scattered across a dozen files.
Commit categories over time, read off the subject line. Fixes carry the accent because that is the series this chart exists to show.
Consistently feature-driven across its history.
Ranked by change-risk, highest first. Priority is a tercile of this repo's own distribution, so a quiet repo still fills its top band.
| # | Commit | Author | When | Lines | Risk | Top driver |
|---|---|---|---|---|---|---|
| 1 | 63fcc20afeat: full graphify mastery — labeled graph, media sources, artifact + update tasks | Raymond Manalotoclaudeassisted | 3w ago | +316.1K -16 | 100%Elevated | more lines added than baseline |
| 2 | f131ea1cStabilize Graphify SDK and source workflows (#280) | Raymond Manaloto | 1d ago | +11.7K -702 | 96%Elevated | more lines added than baseline |
| 3 | fd7e1a5bfeat/187 reextract baseline docs (#234) | Raymond Manalotoclaudeassisted | 6d ago | +20.9K -7.9K | 96%Elevated | more lines added than baseline |
| 4 | bfc6dca8feat: standing agent roster, 12 new sources, and two silent-loss fixes (#209) | Raymond Manaloto | 1w ago | +15.3K -165 | 96%Elevated | more lines added than baseline |
| 5 | d5da30c7chore/post 186 (#197) | Raymond Manalotoclaudeassisted | 1w ago | +21.5K -36 | 96%Elevated | more lines added than baseline |
| 6 | e084df70docs/ray directives 2026 08 02 (#136) | Raymond Manalotoclaudeassisted | 1w ago | +8.9K -2.1K | 96%Elevated | more lines added than baseline |
| 7 | 9698879ddocs(goals): adopt goal+rider pairs, and write the Legible round (#51) | Raymond Manalotoclaudeassisted | 2w ago | +12.4K -2 | 96%Elevated | more lines added than baseline |
| 8 | f7a99804feat(kb): ingest long-running-framework sources + reusable ingestion tasks (#2) | Raymond Manalotoclaudeassisted | 3w ago | +17.2K -14 | 96%Elevated | more lines added than baseline |
| 9 | 509de506feat(kb): ingest+extract Fable-5/orchestrator/advisor/fallback corpus | Raymond Manalotoclaudeassisted | 3w ago | +8.3K -19 | 96%Elevated | more lines added than baseline |
| 10 | 67f7ef0bfeat(graphify): add deterministic 0.9.42 AST baseline (#307) | Raymond Manaloto | 29m ago | +4.0K -27 | 91%Elevated | more lines added than baseline |
| 11 | 459de7adfeat/189 #191 — refuse an undeclared cross-chunk claim; assert the merge arithmetic (#199) | Raymond Manalotoclaudeassisted | 1w ago | +4.4K -1.1K | 91%Elevated | more lines added than baseline |
| 12 | 7ea192e6feat/164 deep round (#180) | Raymond Manalotoclaudeassisted | 1w ago | +3.7K -698 | 91%Elevated | more lines added than baseline |
| 13 | 1bf2e56ffeat(corpus): ingest the skills we RUN, and fix #93 so the sweep can be trusted (#108) | Raymond Manalotoclaudeassisted | 1w ago | +6.7K -5 | 91%Elevated | more lines added than baseline |
| 14 | 43a6b468feat/graph navigation and tool review (#102) | Raymond Manalotoclaudeassisted | 1w ago | +3.9K -12 | 91%Elevated | more lines added than baseline |
| 15 | cfdf6f70feat/scan the commit range for secrets (#67) (#73) | Raymond Manalotoclaudeassisted | 2w ago | +3.6K -35 | 91%Elevated | more lines added than baseline |
| 16 | 3bb49a8cfeat(claude): 22 .claude/rules + md_budget shared engine + agnix wiring (#24) | Raymond Manaloto | 2w ago | +2.9K -5 | 91%Elevated | more lines added than baseline |
| 17 | 2302024cfeat(currency): shared tool-currency engine + graphify pilot (#4) | Raymond Manalotoclaudeassisted | 3w ago | +5.4K -3 | 91%Elevated | more lines added than baseline |
| 18 | dffe600afeat(tooling): track Doppler and offline docs (#278) | Raymond Manaloto | 1d ago | +2.7K -1 | 83%Elevated | more lines added than baseline |
| 19 | f7f1d160feat/section 2 5 stdout sink (#273) | Raymond Manaloto | 3d ago | +2.8K -163 | 83%Elevated | more lines added than baseline |
| 20 | 4596ed55feat/r5 sweep conversion (#271) | Raymond Manalotoclaudeassisted | 4d ago | +1.4K -126 | 83%Elevated | more lines added than baseline |
| 21 | 725044f4feat/disk reclaim (#237) | Raymond Manalotoclaudeassisted | 6d ago | +3.0K -7 | 83%Elevated | more lines added than baseline |
| 22 | 27bf6910chore/graphify 0.9.34 prep (#190) | Raymond Manalotoclaudeassisted | 1w ago | +4.7K -561 | 83%Elevated | more lines added than baseline |
| 23 | a2714cd1chore/resync followups (#173) | Raymond Manalotoclaudeassisted | 1w ago | +2.5K -1 | 83%Elevated | more lines added than baseline |
| 24 | 822c33a3feat/145 kb handoff check (#153) | Raymond Manalotoclaudeassisted | 1w ago | +2.6K -2 | 83%Elevated | more lines added than baseline |
| 25 | 10af8775feat/settled claims (#104) | Raymond Manalotoclaudeassisted | 1w ago | +2.8K -93 | 83%Elevated | more lines added than baseline |
| 26 | dc7de407feat/prove mirror extraction 84 (#92) | Raymond Manalotoclaudeassisted | 2w ago | +2.9K -0 | 83%Elevated | more lines added than baseline |
| 27 | 285aaccfchore/mise currency 2026.7.16 (#77) | Raymond Manalotoclaudeassisted | 2w ago | +2.8K -310 | 83%Elevated | more lines added than baseline |
| 28 | ba9909aefeat/local cross family review (#53) | Raymond Manalotoclaudeassisted | 2w ago | +3.1K -109 | 83%Elevated | more lines added than baseline |
| 29 | 2f54e24bfix/cc doctor judges session path (#47) | Raymond Manalotoclaudeassisted | 2w ago | +2.8K -32 | 83%Elevated | more lines added than baseline |
| 30 | 4c944a3cfeat(kb): ingest Claude-5 context-engineering + Cerebras KB sources, add gap analysis (#11) | Raymond Manaloto | 2w ago | +3.8K -3 | 83%Elevated | more lines added than baseline |
| 31 | f008f1b8feat: knowledge-base — graphify research substrate (mise/hk/uv, Claude-only) | Raymond Manalotoclaudeassisted | 3w ago | +2.3K -0 | 83%Elevated | more lines added than baseline |
| 32 | dfbda98efeat/r5 sweep tranche4 (#272) | Raymond Manalotoclaudeassisted | 3d ago | +1.3K -109 | 74%Elevated | more lines added than baseline |
| 33 | e81ba4a5fix/kb watch replay regression (#258) | Raymond Manalotoclaudeassisted | 5d ago | +2.0K -35 | 74%Elevated | more lines added than baseline |
| 34 | 457a9fbachore/post 221 (#222) | Raymond Manalotoclaudeassisted | 1w ago | +2.0K -8 | 74%Elevated | more lines added than baseline |
| 35 | a2ef5d88fix/181 #182 — restamp after kb-merge, and see a semantically stale derived view (#185) | Raymond Manalotoclaudeassisted | 1w ago | +2.0K -275 | 74%Elevated | more lines added than baseline |
| 36 | 8c4608d1feat/176 hyperedge survival (#183) | Raymond Manalotoclaudeassisted | 1w ago | +1.5K -113 | 74%Elevated | more lines added than baseline |
| 37 | 4a5ab968feat/144 kb session state (#161) | Raymond Manalotoclaudeassisted | 1w ago | +2.0K -61 | 74%Elevated | more lines added than baseline |
| 38 | b72e4413feat/154 extension typo (#159) | Raymond Manalotoclaudeassisted | 1w ago | +1.3K -17 | 74%Elevated | more lines added than baseline |
| 39 | 6584fbd3feat/147 gate claim verify (#156) | Raymond Manalotoclaudeassisted | 1w ago | +1.6K -16 | 74%Elevated | more lines added than baseline |
| 40 | a8ce533bfeat/clear prep and skill sync (#139) | Raymond Manalotoclaudeassisted | 1w ago | +2.0K -14 | 74%Elevated | more lines added than baseline |
| 41 | afba16a8fix/kb serve stdio (#105) | Raymond Manalotoclaudeassisted | 1w ago | +1.5K -1 | 74%Elevated | more lines added than baseline |
| 42 | 1dab0e54feat(evals): golden retrieval set — recall@k per query pair, advisory (#30) | Raymond Manalotoclaudeassisted | 2w ago | +1.2K -35 | 74%Elevated | more lines added than baseline |
| 43 | c1891cb9Add immutable artifact receipt workflow (#286) | Raymond Manaloto | 16h ago | +1.3K -4 | 62%Typical | more lines added than baseline |
| 44 | 10ad4220feat: add reviewed SkillOpt adapter (#285) | Raymond Manaloto | 19h ago | +1.7K -1 | 62%Typical | more lines added than baseline |
| 45 | d60eb958feat/r5 typed error surface (#267) | Raymond Manalotoclaudeassisted | 4d ago | +1.2K -11 | 62%Typical | more lines added than baseline |
| 46 | 918225d6chore/round close 2026 08 09 (#266) | Raymond Manalotoclaudeassisted | 4d ago | +931 -5 | 62%Typical | more lines added than baseline |
| 47 | 6b798265feat: DENY a broad source search until the graph has been queried (#253) (#254) | Raymond Manalotoclaudeassisted | 5d ago | +1.2K -8 | 62%Typical | more lines added than baseline |
| 48 | 7e34dfb7fix: all 15 findings across three cold-review rounds (#245) (#252) | Raymond Manalotoclaudeassisted | 5d ago | +1.1K -84 | 62%Typical | more lines added than baseline |
| 49 | afe28c76feat/198 hyperedge continuity (#230) | Raymond Manalotoclaudeassisted | 6d ago | +1.1K -58 | 62%Typical | more lines added than baseline |
| 50 | a41f0b5cchore/post 220 (#221) | Raymond Manalotoclaudeassisted | 1w ago | +1.4K -134 | 62%Typical | more lines added than baseline |
Two views of the same model: where the cuts fall, and what commit shape lands you above them.
Every scored commit, binned on the raw 0 to 10 score rather than the percentile. Percentile ranks are uniform by construction, so that axis has no shape to draw. The dashed lines are the tercile cuts behind each row's priority pill.
The 138 most recent commits, on their own recency sample rather than the feed above: that defaults to risk-sorted, so reusing it would plot only the top tercile and call it the spread. Big and scattered is what the model penalises. Click a dot to open it.
ray-manaloto/knowledge-base has 139 commits in its history from 1 contributor, the first of them Jul 22, 2026. In the last 90 days 175 files were touched, 760 times in total, most often cli.py. Every commit is scored for change risk from its size, spread and the history of the files it touches.
