[FEATURE]: Tool-calling optimization for Trinity model family #8462

Closed
opened 2026-02-16 18:10:01 -05:00 by yindo · 1 comment
Owner

Originally created by @mariamjabara on GitHub (Feb 3, 2026).

Originally assigned to: @thdxr on GitHub.

Feature hasn't been suggested before.

  • I have verified this feature I'm about to request hasn't been suggested before.

Describe the enhancement you want to request

Add a dedicated system prompt for the Trinity model family to enforce:

  • one tool per message
  • Use of the question tool when the request is vague
  • Avoid repeating the same tool with the same params

Benefits: improved tool-calling reliability and a smoother experience when using Trinity models.

Originally created by @mariamjabara on GitHub (Feb 3, 2026). Originally assigned to: @thdxr on GitHub. ### Feature hasn't been suggested before. - [x] I have verified this feature I'm about to request hasn't been suggested before. ### Describe the enhancement you want to request Add a dedicated system prompt for the Trinity model family to enforce: - one tool per message - Use of the question tool when the request is vague - Avoid repeating the same tool with the same params Benefits: improved tool-calling reliability and a smoother experience when using Trinity models.
yindo added the discussion label 2026-02-16 18:10:01 -05:00
yindo closed this issue 2026-02-16 18:10:01 -05:00
Author
Owner

@github-actions[bot] commented on GitHub (Feb 3, 2026):

This issue might be a duplicate or closely related to existing issues addressing model-specific system prompt optimization. Please check:

  • #11730: Add GLM specific system prompt support - directly parallel feature for a different model family
  • #5728: Ability to override tool prompts - create model toolsets
  • #9065: Add model identity info to system prompt
  • #3195: Dynamic system prompts for custom agents
  • #11053: Kimi k2.5 doesn't use the question tool
  • #9525: Question tool is way too eager
  • #3458: Duplicate Executions in Kimi K2 Model
  • #234: Tool Calling Issues with Open Source Models in OpenCode
  • #10927: Default system prompt negatively impacts high reasoning models
  • #9570: Support Gemini API by trimming/adapting tool descriptions

Feel free to ignore if your request addresses a specific gap not covered in these issues.

@github-actions[bot] commented on GitHub (Feb 3, 2026): This issue might be a duplicate or closely related to existing issues addressing model-specific system prompt optimization. Please check: - #11730: Add GLM specific system prompt support - directly parallel feature for a different model family - #5728: Ability to override tool prompts - create model toolsets - #9065: Add model identity info to system prompt - #3195: Dynamic system prompts for custom agents - #11053: Kimi k2.5 doesn't use the question tool - #9525: Question tool is way too eager - #3458: Duplicate Executions in Kimi K2 Model - #234: Tool Calling Issues with Open Source Models in OpenCode - #10927: Default system prompt negatively impacts high reasoning models - #9570: Support Gemini API by trimming/adapting tool descriptions Feel free to ignore if your request addresses a specific gap not covered in these issues.
Sign in to join this conversation.
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: anomalyco/opencode#8462