No UTF-8 byte-order mark

Reject text files that begin with the UTF-8 byte-order mark (EF BB BF, U+FEFF). The BOM is unnecessary for UTF-8 and silently breaks shell scripts (interpreter line not recognised), shebang handling, JSON parsers, and many migration tools.

CSV and TSV files are exempt - Excel writes a BOM by default and the files often round-trip through it.

Fix#

Strip the BOM:

sed -i '1s/^\xEF\xBB\xBF//' <file>

or in editors with “save without BOM” / “encoding: UTF-8 (no BOM)”.

If a specific file legitimately needs a BOM (rare Windows tooling), suppress at the file level with # appframes:disable encoding/no-bom.