UTF8 Encoding

Question

hello, I would like that on certain script I could be able to save the file in utf8 with BOM instead of utf8 only. how can I do it please?

u want opencode to create non utf8 files, is that correct (or the option to) via BOM

Yes because when an AI makes a modification to a script, this one, opencode modifies the encoding to UTF8, and I would like this to keep the encoding of the file which it was in UTF8 with Bom

gotcha

I am interested in that feature too. In my case, I am working on a project encoded with ISO-8859-1. Every time a file is edited, it transforms the file into UTF-8. I tried to turn the formatter option to false (https://opencode.ai/docs/formatters/#disabling-formatters)

Any progress?

This issue is actually bigger than it suggests: OpenCode corrupts PowerShell files

OpenCode strips UTF-8 BOM when editing files, which breaks PowerShell scripts with international characters (å, ä, ö, etc.). PowerShell requires UTF-8 BOM to correctly interpret non-ASCII characters.

Reproduction:

example.ps1 (UTF-8 with BOM)
Write-Host "Börjar bearbetning..."

Before OpenCode edit: file example.ps1 → "UTF-8 (with BOM)" ✓ works
After OpenCode edit: file example.ps1 → "UTF-8" (no BOM) ✗ shows gibberish in PS 5.1

Any edit to a PowerShell file removes the BOM and corrupts the script, making OpenCode unusable for PowerShell development in non-English environments.

Expected:

Preserve UTF-8 BOM for .ps1/.psm1/.psd1 files.

Further investigation shows:

The issue is compounded by saving files with LF line endings and certain PowerShell syntax patterns in PowerShell 5.1.

While simple scripts with LF line endings may parse correctly, PowerShell 5.1 fails to parse files with LF line endings when they contain method calls with quoted string arguments, such as .Trim('"').
Minimal reproduction case:

test-complex.ps1 (saved with LF line endings, UTF-8 without BOM)

	$Line = "url;[email protected];high"
	$parts = $Line -split ';'
	$triggerLevel = $parts[2].Trim('"')
	Write-Host "Result: $triggerLevel"

PowerShell 5.1 output:

At test-complex.ps1:3 char:34
+ $triggerLevel = $parts[2].Trim('"')
+                                  ~~
The string is missing the terminator: '.

Solution: Converting the file to CRLF line endings (with or without UTF-8 BOM) resolves the parsing error completely. PowerShell 7+ does not exhibit this issue and handles LF line endings correctly.

Note on tooling: Standard Unix tools like sed, awk, and Git Bash (MINGW) typically convert files to LF line endings, making them incompatible with PowerShell 5.1 for scripts using this syntax pattern. When modifying PowerShell scripts in mixed Unix/Windows environments, the user must verify line ending format manually before deployment to Windows PowerShell 5.1 environments.

Root cause: PowerShell 5.1's parser incorrectly handles quoted strings as method arguments when preceded by LF-only line terminators, causing it to misinterpret the quote termination.

Why not pay attention to it? This is a big problem

I am also experiencing this issue. More specifically I have Windows-1252 encoded files, that keep being changed to UTF-8, breaking international characters, thus introducing compilation erros

Same issue with ISO encoding.

Automated: closing question issues.

GitHub Issues are for bug reports and feature requests only — questions / support are no longer tracked here.

For help & questions: https://discord.gg/opencode
Billing issues with Opencode Go / Zen: [email protected]

If this is a bug or feature request, please open a new issue with the appropriate template.

@rekram1-node

@rekram1-node Are you deliberately misinterpreting the issue and actively trying to kill the product and its community, because this is how you do it. Get your act together. Please do not interpret politeness or incompleteness as "a non issue" when there is clear evidence of such! The bug here is that Open code is changing the file format for edited files, destroying working code. since the write-file-format is UTF8 without BOM only.

Navigation Menu

Repository navigation

UTF8 Encoding #3342

Question

This issue is actually bigger than it suggests: OpenCode corrupts PowerShell files

Reproduction:

Expected:

Further investigation shows:

test-complex.ps1 (saved with LF line endings, UTF-8 without BOM)

PowerShell 5.1 output:

Metadata

Assignees

Labels

Type

Projects

Milestone

Relationships

Development

UTF8 Encoding #3342

Description

Question

Activity

rekram1-node commented on Oct 21, 2025

LexBoosT commented on Oct 22, 2025

rekram1-node commented on Oct 22, 2025

renanfranca commented on Jan 22, 2026

13640535881 commented on Feb 12, 2026

Mikael-Gustavsson commented on Feb 16, 2026

This issue is actually bigger than it suggests: OpenCode corrupts PowerShell files

Reproduction:

Expected:

Mikael-Gustavsson commented on Feb 17, 2026

Further investigation shows:

test-complex.ps1 (saved with LF line endings, UTF-8 without BOM)

PowerShell 5.1 output:

13640535881 commented on Mar 11, 2026

Anantonon commented on Mar 23, 2026

adpeyre commented on May 5, 2026

rekram1-node commented on Jun 23, 2026

Mikael-Gustavsson commented on Jun 23, 2026

Metadata

Metadata

Assignees

Labels

Type

Projects

Milestone

Relationships

Development

Issue actions