Skip to content

merge dev into master #264

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom
Merged
Tarquinen merged 4 commits into from
Jan 15, 2026
Merged

merge dev into master #264

Tarquinen merged 4 commits into from
Jan 15, 2026

Conversation

@Tarquinen
Copy link
Collaborator

@Tarquinen Tarquinen commented Jan 15, 2026 •
edited
Loading

  • Simplify injection timing, instead of checking provider/model, universally only inject if previous message is not user. This complies with many providers requiring first assistant turn in reasoning models to have reasoning block
  • Change gemini injection to use text part instead of tool part in assistant injections as not all oauth plugins support this properly

Tarquinen and others added 4 commits January 15, 2026 12:17
@Tarquinen
fix: simplify injection timing to wait for assistant turn
e3d74f3 
Replace provider-specific checks (GitHub Copilot, Anthropic reasoning detection) with a universal rule: never inject immediately following a user message. This avoids interfering with model reasoning/thinking phases across all providers.
@Tarquinen
Merge pull request #263 from Opencode-DCP/fix/simplify-injection-timing
018a0d6 
fix: simplify injection timing to wait for assistant turn
@Tarquinen
fix: use text injection for Gemini models to avoid thought signature …
a2ae844 
…errors

Gemini 3+ models have strict validation requiring thoughtSignature on functionCall parts. When the plugin injects synthetic assistant messages with tool parts, providers without robust signature handling will fail with 400 errors.

This change detects Gemini models and injects as text parts instead of tool parts, avoiding the thought signature requirement entirely.
@Tarquinen
Merge pull request #265 from Opencode-DCP/fix/gemini-thought-signatur…
bea6a42 
…e-injection

fix: use text injection for Gemini models to avoid thought signature errors
@Tarquinen Tarquinen merged commit bdc5f78 into master Jan 15, 2026
1 check passed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Reviewers

No reviews

Assignees

No one assigned

Labels

None yet

Projects

None yet

Milestone

No milestone

Development

Successfully merging this pull request may close these issues.

2 participants

@Tarquinen