Answer principles round 1: 0 🔴, 2 🟡 (0 misread); run.sh reads the diff whatever the git config, item 58 ports it to Python, the worker count sits beside the memory, the README says what a run reports

README.md

@@ -733,8 +733,8 @@ Both commands need only Docker, git and bash 5.
 
 - `./ci.sh` runs the gate: every check CI runs.
 - `./mutation-tests/run.sh [<base>]` mutation-tests the source lines changed since `<base>`,
-  `origin/main` by default, running each mutant, in a 6 GB container, against only the test files
-  that reach it.
+  `origin/main` by default, running each mutant against only the test files that reach it. It lists
+  the mutants that survived, and fails only when a mutant's test run crashed.
 
 ## Links
 

mutation-tests/run.sh

@@ -5,8 +5,8 @@ source ./docker-runner.sh
 
 base=$(git merge-base "${1:-origin/main}" HEAD)
 # Each hunk's added lines as path:first-last, in the working tree's numbering; a hunk that only deletes has none to mutate.
-mutate=$(git diff -U0 --no-color "$base" -- 'src/*.ts' ':!src/*.test.ts' ':!src/conformance/property-harness.ts' | awk '
-  /^\+\+\+ / { path = substr($2, 3) }
+mutate=$(git diff --no-ext-diff --no-prefix -U0 --no-color "$base" -- 'src/*.ts' ':!src/*.test.ts' ':!src/conformance/property-harness.ts' | awk '
+  /^\+\+\+ / { path = $2 }
   /^@@ / {
     split($3, added, ",")
     first = substr(added[1], 2)
@@ -19,4 +19,5 @@ if [ -z "$mutate" ]; then
 fi
 
 leg "install ($node_image)" in_image "$node_image" sh -c 'npm ci --no-audit --no-fund && npm ci --no-audit --no-fund --prefix mutation-tests'
-in_image_memory=6g leg "mutation tests ($node_image)" in_image "$node_image" sh -c 'mutation-tests/node_modules/.bin/stryker run mutation-tests/stryker.config.mjs --mutate "$1" && node mutation-tests/check-runtime-errors.mjs' sh "$mutate"
+# A test file peaks near 2.6 GB: past four workers 6 GB kills some, and a killed worker's mutant reads as a RuntimeError, which fails the run.
+in_image_memory=6g leg "mutation tests ($node_image)" in_image "$node_image" sh -c 'mutation-tests/node_modules/.bin/stryker run mutation-tests/stryker.config.mjs --concurrency 4 --mutate "$1" && node mutation-tests/check-runtime-errors.mjs' sh "$mutate"

mutation-tests/stryker.config.mjs

@@ -1,6 +1,4 @@
 export default {
-  // A test file peaks near 2.6 GB: past four workers run.sh's 6 GB container kills some, and a killed worker's mutant reads as a RuntimeError, which fails the run.
-  concurrency: 4,
   coverageAnalysis: 'perTest',
   ignorePatterns: ['/browser-tests', '/converter-comparison', '/dist', '/mutation-tests', '/package-tests'],
   jsonReporter: { fileName: '/tmp/mutation-report.json' },

todo.md

@@ -56,7 +56,7 @@
 | 56 | 0.5.0 |  |  | **Give each piece of `blocks.ts`'s block-walk state and `inline-content.ts`'s `Scan` one owner that returns what it changes.** | 3 | 1 | 2 | 4 | 5 | code health by 5 | 13.1 |
 | 118 | 0.5.0 |  |  | **Spell a portable link's title through `oneLine`.** | 1 | 1 | 1 | 1 | 1 | code health by 1 | 10.5 |
 | 82 | 0.5.0 |  |  | **Give the directive tests and fixtures random UUID v7 ids in place of `a-1`, `a`, `x` and `h`.** | 2 | 1 | 1 | 2 | 3 | code health by 1 | 6.5 |
-| 58 | 0.5.0 |  |  | **Port `ci.sh`, `publish.sh` and `docker-runner.sh` to standalone Python scripts.** | 4 | 2 | 2 | 6 | 6 | code health by 3 | 4.4 |
+| 58 | 0.5.0 |  |  | **Port `ci.sh`, `publish.sh`, `docker-runner.sh` and `mutation-tests/run.sh` to standalone Python scripts.** | 4 | 2 | 2 | 6 | 6 | code health by 3 | 4.4 |
 | 9 | 0.5.0 |  |  | **Ship an online sandbox: a web page with two textboxes converting between ADF and markdown on the library's browser build.** | 3 | 2 | 1 | 4 | 6 | 3 by 6 | 2.1 |
 | 34 | 0.5.0 |  |  | **Read emphasis flanking by the whole character beside an astral symbol.** | 2 | 1 | 2 | 3 | 3 | 4 by 1 | 1.5 |
 | 74 | 0.5.0 |  | owner | **Keep the release path publishing after npm retires bypass-2FA tokens.** | 3 | 2 | 4 | 7 | 4 |  | 0.0 |
@@ -466,10 +466,11 @@ Found by the principles audit of item 99, 2026-10-14.
 Technical principle: tests use random UUID v7 ids; `portable-markdown-to-adf.test.ts` already does.
 Found by the technical-principles audit of item 78, 2026-10-05.
 
-### 58. Port `ci.sh`, `publish.sh` and `docker-runner.sh` to standalone Python scripts.
+### 58. Port `ci.sh`, `publish.sh`, `docker-runner.sh` and `mutation-tests/run.sh` to standalone Python scripts.
 
 Technical principle "Prefer a standalone Python script over a shell script": `AGENTS.md` §3 teaches
-bash footguns (`&&` chaining under `||`, `tee /dev/stderr`) the port removes.
+bash footguns (`&&` chaining under `||`, `tee /dev/stderr`) the port removes. The port tests
+`run.sh`'s turning of diff hunks into mutated line ranges, which a wrong range would pass silently.
 
 ### 34. Read emphasis flanking by the whole character beside an astral symbol.