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
133 commits sit in this repo's top risk tercile, which is 34% of the 387scored. 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 9.0 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.
How this repo's commit mix shifts over time.
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 | 6b533157fix: auto compaction threshold default 200k tokens | Yeachan-Heo | 5mo ago | +31.1K -0 | 100thElevated | more lines added than baseline |
| 2 | 44e47580feat: Rust port of Claude Code CLI | Yeachan-Heo | 5mo ago | +8.1K -0 | 100thElevated | more lines added than baseline |
| 3 | 7eb63307feat: Rust port of Claw Code CLI | Yeachan-Heo | 5mo ago | +8.1K -0 | 100thElevated | more lines added than baseline |
| 4 | 3faf8dd3feat: make rusty-claude-cli usable end-to-end | Yeachan-Heo | 5mo ago | +2.8K -86 | 99thElevated | more lines added than baseline |
| 5 | 52af1f22feat: make claw-cli usable end-to-end | Yeachan-Heo | 5mo ago | +2.8K -86 | 99thElevated | more lines added than baseline |
| 6 | 45055655feat: merge 2nd round from all rcc/* sessions | Yeachan-Heo | 5mo ago | +2.4K -3.6K | 99thElevated | more lines added than baseline |
| 7 | 334d1854feat: merge 2nd round from all rcc/* sessions | Yeachan-Heo | 5mo ago | +2.4K -3.6K | 99thElevated | more lines added than baseline |
| 8 | 618a79a9feat: ultraclaw session outputs — registry tests, MCP bridge, PARITY.md, cleanup | Jobdori | 4mo ago | +1.7K -204 | 98thElevated | more lines added than baseline |
| 9 | 799ee3a4wip: plugins progress | Yeachan-Heo | 5mo ago | +2.5K -21 | 98thElevated | more lines added than baseline |
| 10 | 2a0f4b67feat: provider abstraction layer + Grok API support | Yeachan-Heo | 5mo ago | +1.5K -997 | 98thElevated | more lines added than baseline |
| 11 | 5654efb7feat: provider abstraction layer + Grok API support | Yeachan-Heo | 5mo ago | +1.5K -997 | 98thElevated | more lines added than baseline |
| 12 | 507c2460Make the repository's primary source tree genuinely Python | instructkr | 5mo ago | +4.0K -0 | 98thElevated | more lines added than baseline |
| 13 | f76311f9Prevent worker prompts from outrunning boot readiness | Yeachan-Heo | 4mo ago | +1.1K -4 | 96thElevated | more lines added than baseline |
| 14 | b3fe0575Close the MCP lifecycle gap from config to runtime tool execution | Yeachan-Heo | 4mo ago | +1.2K -105 | 96thElevated | more lines added than baseline |
| 15 | 1cfd78acfeat: bash validation module + output truncation parity | Jobdori | 4mo ago | +1.1K -2 | 96thElevated | more lines added than baseline |
| 16 | ee0c4cd0feat: plugin subsystem progress | Yeachan-Heo | 5mo ago | +1.2K -274 | 96thElevated | more lines added than baseline |
| 17 | eb89fc95wip: hook-pipeline progress | Yeachan-Heo | 5mo ago | +1.2K -138 | 96thElevated | more lines added than baseline |
| 18 | c26797d9wip: hook-pipeline progress | Yeachan-Heo | 5mo ago | +1.2K -138 | 96thElevated | more lines added than baseline |
| 19 | cbc0a830auto: save WIP progress from rcc session | Yeachan-Heo | 5mo ago | +1.2K -21 | 96thElevated | more lines added than baseline |
| 20 | 8eb40bc6auto: save WIP progress from rcc session | Yeachan-Heo | 5mo ago | +1.2K -21 | 96thElevated | more lines added than baseline |
| 21 | 01bf54adRewriting Project Claw Code - Python port with Rust on the way | instructkr | 5mo ago | +1.2K -111 | 96thElevated | more lines added than baseline |
| 22 | 8a9ea167feat(mcp+lifecycle): MCP degraded-startup reporting, lane event schema, lane completion hardening | Yeachan-Heo | 4mo ago | +804 -184 | 94thElevated | more lines added than baseline |
| 23 | 85c5b0e0Expand parity harness coverage before behavioral drift lands | Yeachan-Heo | 4mo ago | +1.2K -97 | 94thElevated | more lines added than baseline |
| 24 | 3b18ce9ffeat(mcp): add toolCallTimeoutMs, timeout/reconnect/error handling | YeonGyu-Kim | 5mo ago | +861 -130 | 94thElevated | more lines added than baseline |
| 25 | 5d14ff1dfeat: plugin subsystem — loader, hooks, tools, bundled, CLI | Yeachan-Heo | 5mo ago | +965 -178 | 94thElevated | more lines added than baseline |
| 26 | 51707183wip: telemetry progress | Yeachan-Heo | 5mo ago | +1.1K -70 | 94thElevated | more lines added than baseline |
| 27 | 0cf2204dwip: cache-tracking progress | Yeachan-Heo | 5mo ago | +916 -5 | 94thElevated | more lines added than baseline |
| 28 | 992681c4Prevent long sessions from stalling and expose the requested internal command surface | Yeachan-Heo | 5mo ago | +733 -9 | 94thElevated | more lines added than baseline |
| 29 | 13015f64feat(runtime): hardened MCP lifecycle with phase tracking and degraded-mode reporting | Jobdori | 4mo ago | +820 -30 | 89thElevated | more lines added than baseline |
| 30 | c2f1304aLock down CLI-to-mock behavioral parity for Anthropic flows | Yeachan-Heo | 4mo ago | +1.1K -2 | 89thElevated | more lines added than baseline |
| 31 | 03bd7f05feat: add 40 slash commands — command surface 67/141 | Jobdori | 4mo ago | +652 -15 | 89thElevated | more lines added than baseline |
| 32 | d88144d4feat(commands): slash-command validation, help formatting, CLI wiring | YeonGyu-Kim | 5mo ago | +612 -35 | 89thElevated | more lines added than baseline |
| 33 | 731ba984feat(commands): add slash command argument validation with clear error messages | YeonGyu-Kim | 5mo ago | +664 -167 | 89thElevated | more lines added than baseline |
| 34 | c8060355wip: jsonl-session progress | Yeachan-Heo | 5mo ago | +723 -103 | 89thElevated | more lines added than baseline |
| 35 | ac6c5d00Enable Claude-compatible tool hooks in the Rust runtime | Yeachan-Heo | 5mo ago | +653 -13 | 89thElevated | more lines added than baseline |
| 36 | b40fb0c4Enable compatible tool hooks in the Rust runtime | Yeachan-Heo | 5mo ago | +653 -13 | 89thElevated | more lines added than baseline |
| 37 | 705c6225Improve terminal output so Rust CLI renders readable rich responses | Yeachan-Heo | 5mo ago | +559 -109 | 89thElevated | more lines added than baseline |
| 38 | d0327f65Improve terminal output so Rust CLI renders readable rich responses | Yeachan-Heo | 5mo ago | +559 -109 | 89thElevated | more lines added than baseline |
| 39 | 387a8bb1feat: git integration, sandbox isolation, init command (merged from rcc branches) | Yeachan-Heo | 5mo ago | +772 -1.3K | 89thElevated | more lines added than baseline |
| 40 | 98264aa3feat: git integration, sandbox isolation, init command (merged from rcc branches) | Yeachan-Heo | 5mo ago | +772 -1.3K | 89thElevated | more lines added than baseline |
| 41 | 2d09bf99Make sandbox isolation behavior explicit and inspectable | Yeachan-Heo | 5mo ago | +720 -22 | 89thElevated | more lines added than baseline |
| 42 | b200198dMake local environment failures diagnosable from the CLI | Yeachan-Heo | 5mo ago | +728 -15 | 89thElevated | more lines added than baseline |
| 43 | d8c6a300Make local environment failures diagnosable from the CLI | Yeachan-Heo | 5mo ago | +728 -15 | 89thElevated | more lines added than baseline |
| 44 | cd01d0e3Honor Claude config defaults across runtime sessions | Yeachan-Heo | 5mo ago | +630 -60 | 89thElevated | more lines added than baseline |
| 45 | d3ab7d9cHonor config defaults across runtime sessions | Yeachan-Heo | 5mo ago | +630 -60 | 89thElevated | more lines added than baseline |
| 46 | 8d4a739cMake the REPL resilient enough for real interactive workflows | Yeachan-Heo | 5mo ago | +610 -88 | 89thElevated | more lines added than baseline |
| 47 | 1104da21Make the REPL resilient enough for real interactive workflows | Yeachan-Heo | 5mo ago | +610 -88 | 89thElevated | more lines added than baseline |
| 48 | 32e89df6Enable Claude OAuth login without requiring API keys | Yeachan-Heo | 5mo ago | +753 -27 | 89thElevated | more lines added than baseline |
| 49 | 568f5f90Enable OAuth login without requiring API keys | Yeachan-Heo | 5mo ago | +753 -27 | 89thElevated | more lines added than baseline |
| 50 | 1e5002b5Add MCP server orchestration so configured stdio tools can be discovered and called | Yeachan-Heo | 5mo ago | +767 -7 | 89thElevated | 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 200 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.
Repowise tracks change history across 143 files in HexSleeves/claw-code-parity. No file in the repository changed in the last 90 days. Every commit is scored for change risk from its size, spread and the history of the files it touches.
