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
15 commits sit in this repo's top risk tercile, which is 33% of the 46scored. 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 8.5 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 | Review priority | Top driver |
|---|---|---|---|---|---|---|
| 1 | 6a555b4fdocs(slides): deck plans, speaker notes, and master spec for Claude Design | Russ Milesclaudeassisted | 1mo ago | +3.1K -2 | 99thElevated | more lines added than baseline |
| 2 | d7bac622docs: add Diátaxis documentation site (MkDocs Material + GitHub Pages) | Russ Milesclaudeassisted | 2mo ago | +2.3K -97 | 97thElevated | more lines added than baseline |
| 3 | d7f6ae28feat: baseline ConfPlanner agent | Russ Milesclaudeassisted | 2mo ago | +1.5K -0 | 95thElevated | more lines added than baseline |
| 4 | e504eeadInitial commit | Russ Miles | 2mo ago | +1.7K -0 | 92thElevated | more lines added than baseline |
| 5 | 256f028ffeat(lab5): NetworkingPlan + planNetworking via a new agent | Russ Milesclaudeassisted | 2mo ago | +398 -191 | 90thElevated | more lines added than baseline |
| 6 | 84ce905cfeat: make the mock profile actually mock (#5) (#23) | Russ Milesclaudeassisted | 1mo ago | +421 -2 | 87thElevated | more lines added than baseline |
| 7 | 74ac290bdocs: workshop README, SETUP, HABITS, slides outline, harness files | Russ Milesclaudeassisted | 2mo ago | +484 -156 | 87thElevated | more lines added than baseline |
| 8 | 9c3ab518docs(quizzes): add a short quiz per section; tutor administers and discusses them (#34) | Russ Milesclaudeassisted | 1mo ago | +326 -4 | 83thElevated | more lines added than baseline |
| 9 | ca195674feat(tutor): add guided workshop tutor agent with solo + in-room modes (#29) | Russ Milesclaudeassisted | 1mo ago | +331 -0 | 83thElevated | more lines added than baseline |
| 10 | fdea92a1feat(eval): add the eval lane — sampled LLM-as-judge for schedule quality (#33) | Russ Milesclaudeassisted | 1mo ago | +383 -10 | 79thElevated | more lines added than baseline |
| 11 | 4ad4ce19docs(slides): add habitat-thinking cheatsheet — the amber thread mapped by module (#35) | Russ Milesclaudeassisted | 1mo ago | +222 -0 | 75thElevated | more lines added than baseline |
| 12 | 8dc48a70docs: learner-walkthrough audit v5 — slides sync + Lab 4 keyless-path residual (#28) | Russ Milesclaudeassisted | 1mo ago | +221 -44 | 75thElevated | more lines added than baseline |
| 13 | 882a8610feat(observability): AgenticEventListener flight-recorder exercise for Lab 4 | Russ Milesclaudeassisted | 1mo ago | +258 -10 | 75thElevated | more lines added than baseline |
| 14 | 0412797cfeat(lab3): noDoubleBooking invariant + precondition + budget + secured action | Russ Milesclaudeassisted | 2mo ago | +205 -17 | 71thElevated | more lines added than baseline |
| 15 | 2dff312adocs: use SPRING_PROFILES_ACTIVE for profile activation (#4) (#17) | Russ Milesclaudeassisted | 1mo ago | +346 -10 | 68thElevated | more lines added than baseline |
| 16 | 03931266docs(labs): warn about reference-only test changes; label key-only steps (#9, #16) (#20) | Russ Milesclaudeassisted | 1mo ago | +96 -22 | 65thTypical | more lines added than baseline |
| 17 | 28905151feat(guardrails): content guardrail — the fifth shape, on main | Russ Milesclaudeassisted | 1mo ago | +155 -0 | 65thTypical | more lines added than baseline |
| 18 | 9a3a1815feat(quiz): add the /quiz room-quiz command (#38) | Russ Milesclaudeassisted | 1mo ago | +107 -0 | 62thTypical | more lines added than baseline |
| 19 | 64ea4de6docs: learner-walkthrough audit v2 — worksheet & doc fixes (#24) | Russ Milesclaudeassisted | 1mo ago | +102 -28 | 60thTypical | more lines added than baseline |
| 20 | 1ba3e10cdocs(labs): post-0.5.0 going-further extensions + lab7 stretch worksheet | Russ Milesclaudeassisted | 1mo ago | +120 -2 | 57thTypical | more lines added than baseline |
| 21 | 8b88c8fefeat(lab2): researchSessions action; assemble consumes ResearchedSessions | Russ Milesclaudeassisted | 2mo ago | +86 -8 | 57thTypical | more lines added than baseline |
| 22 | 39573b14chore(lab3): scaffold + worksheet | Russ Milesclaudeassisted | 2mo ago | +97 -0 | 53thTypical | more lines added than baseline |
| 23 | aebf1383docs: learner-walkthrough audit v3 — Lab 7 recipe + residual worksheet nits (#25) | Russ Milesclaudeassisted | 1mo ago | +70 -22 | 50thTypical | more lines added than baseline |
| 24 | 48bcb4fefeat(lab1): avoidTopics field + PromptContributor on AttendeeProfile | Russ Milesclaudeassisted | 2mo ago | +67 -15 | 50thTypical | more lines added than baseline |
| 25 | 2edd91e2docs(how-to): add "route a step to a local model" guide; wire into docs, slides, and the tutor (#31) | Russ Milesclaudeassisted | 1mo ago | +151 -5 | 47thTypical | more lines added than baseline |
| 26 | 804a2a3dfix(lab4): inject deliberate fault for debugging lab | Russ Milesclaudeassisted | 2mo ago | +74 -1 | 45thTypical | more lines added than baseline |
| 27 | 74078b7ddocs: fix three honesty oversells surfaced by an adversarial skeptic audit (#32) | Russ Milesclaudeassisted | 1mo ago | +74 -12 | 40thTypical | more lines added than baseline |
| 28 | d6b04323docs: risk register in the docs site | Russ Milesclaudeassisted | 1mo ago | +86 -0 | 40thTypical | more lines added than baseline |
| 29 | 8ac284fachore(lab5): scaffold + worksheet | Russ Milesclaudeassisted | 2mo ago | +72 -0 | 40thTypical | more lines added than baseline |
| 30 | bd8157a6docs: accuracy fixes + Track C prereqs + Lab 7 discoverability (#14, #11, #13) (#19) | Russ Milesclaudeassisted | 1mo ago | +74 -16 | 34thTypical | more lines added than baseline |
| 31 | e0e8e5adfix(lab4): repair the broken plan (+ regression test) | Russ Milesclaudeassisted | 2mo ago | +64 -6 | 34thTypical | more lines added than baseline |
| 32 | 48589a41chore(lab2): scaffold + worksheet | Russ Milesclaudeassisted | 2mo ago | +68 -0 | 34thTypical | more lines added than baseline |
| 33 | 34fdd5e4feat: load .env and put providers always on the classpath (#3) (#22) | Russ Milesclaudeassisted | 1mo ago | +62 -43 | 28thBelow typical | more lines added than baseline |
| 34 | fcfe672bchore(lab6): scaffold + worksheet | Russ Milesclaudeassisted | 2mo ago | +62 -0 | 28thBelow typical | more lines added than baseline |
| 35 | 86282fecdocs: fix clone-dead reference/ citations and Lab 4's thin keyless Zipkin step (#37) | Russ Milesclaudeassisted | 1mo ago | +56 -26 | 24thBelow typical | more lines added than baseline |
| 36 | c884a666feat(lab6): per-action model routing + MODEL_ROUTING.md | Russ Milesclaudeassisted | 2mo ago | +47 -7 | 24thBelow typical | more lines added than baseline |
| 37 | 5b83a1a6docs: tighten Diátaxis quadrant purity (advocatus-diaboli follow-ups) | Russ Milesclaudeassisted | 2mo ago | +27 -36 | 21thBelow typical | more lines added than baseline |
| 38 | 519592dddocs: learner-walkthrough audit v4 — fix the keyless run path (plan, not x) (#27) | Russ Milesclaudeassisted | 1mo ago | +35 -12 | 17thBelow typical | more lines added than baseline |
| 39 | c43ff874docs: fix routing-table contradiction and add an Anthropic mapping (#12) (#21) | Russ Milesclaudeassisted | 1mo ago | +34 -11 | 17thBelow typical | more lines added than baseline |
| 40 | 9548407bchore: scaffold ConfPlanner from java-agent-template, pin Embabel 0.5.0 | Russ Milesclaudeassisted | 2mo ago | +28 -365 | 14thBelow typical | more lines added than baseline |
| 41 | 5e04c101fix(tutor): generalize keyless-observation guidance; flag Lab 6 as the key-needed exception (#30) | Russ Milesclaudeassisted | 1mo ago | +20 -0 | 11thBelow typical | more lines added than baseline |
| 42 | e6afb5d9docs(lab7): make the RAG recipe runtime-verified, not just compile-verified (#26) | Russ Milesclaudeassisted | 1mo ago | +19 -10 | 11thBelow typical | more lines added than baseline |
| 43 | 34ebee65docs(tutor): weave the habitat-thinking cheatsheet (amber thread) into the tutor (#36) | Russ Milesclaudeassisted | 1mo ago | +12 -0 | 8thBelow typical | more focused than baseline |
| 44 | 09e3f114chore: update repo references after transfer to Habitat-Thinking | Russ Milesclaudeassisted | 2mo ago | +12 -12 | 5thBelow typical | more scattered than baseline |
| 45 | 069f14d2chore: shed Embabel template residue and stop tracking runtime artifacts (#15) (#18) | Russ Milesclaudeassisted | 1mo ago | +10 -475 | 3thBelow typical | more files than baseline |
| 46 | b36fa792chore: ignore local reference/ source materials | Russ Milesclaudeassisted | 1mo ago | +5 -0 | 1thBelow typical | fewer 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 46 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.
jedwardsche/hands-on-agent-development has 47 commits in its history from 1 contributor, the first of them Jun 24, 2026. In the last 90 days 35 files were touched, 60 times in total, most often ConfPlannerAgent.java. Every commit is scored for change risk from its size, spread and the history of the files it touches.
