Tracker: `.agents/skills/comment-and-doc-style/references/line-endings.md`

[ptr727]

Tracker for .agents/skills/comment-and-doc-style/references/line-endings.md. Filed under #1315's roll-up of the open prose backlog: every pre-existing finding a whole-unit read raises against this file is listed here once, one line each, naming the unit and the pull request it came from, so the finding is not lost and no pull request carries it as a block. A new pre-existing finding on this file is appended here rather than filed as its own issue. The file's first whole-unit reads were on #1388, which moved the workflow YAML LF pin into "Operational (config) repos".

1. "adds narrow matching LF overrides in both files only for scripts it executes on POSIX" contradicts the mixed-consumer fallback in "Operational (config) repos", which is stated direction-agnostically, so a CRLF-native repo with a Linux-consumed subtree is told by one sentence to pin it LF and by this one that no such override exists. [line-endings.md > Scripts and extensionless executables] (Cut WORKFLOW.md Section 2 to a Pointer at Workflow YAML Conventions #1388)
2. "Must be LF." is unqualified for every script and extensionless executable, while the reason given applies to POSIX execution and the third sentence limits a CRLF-native repo's LF pins to scripts it executes on POSIX, and the unit does not say which wins for a Windows-executed script. [line-endings.md > Scripts and extensionless executables] (Cut WORKFLOW.md Section 2 to a Pointer at Workflow YAML Conventions #1388)
3. The byte-preserve case spells out the .editorconfig half property by property and never gives the .gitattributes half, so under * text=auto eol=lf a reader who follows the unit literally leaves git normalizing every detected-text file in that directory. [line-endings.md > Scripts and extensionless executables] (Cut WORKFLOW.md Section 2 to a Pointer at Workflow YAML Conventions #1388)
4. "an ending the [*] default no longer supplies" asserts a history a first-time reader cannot verify from the unit, stale phrasing from a default change the unit does not describe. [line-endings.md > Scripts and extensionless executables] (Cut WORKFLOW.md Section 2 to a Pointer at Workflow YAML Conventions #1388)
5. The byte-preserve recipe's insert_final_newline = false is not a no-op, since the EditorConfig spec defines false as ensuring the file does not end with a newline, so an editor implementing it literally strips an existing trailing newline in exactly the directory the sentence says must keep its bytes, where unset is the value the same sentence applies to the other two properties. [line-endings.md > Scripts and extensionless executables] (Cut WORKFLOW.md Section 2 to a Pointer at Workflow YAML Conventions #1388)