[PR #14] [MERGED] fix: pass selected model to backend when invoking agent #27

New Issue

2026-02-16T09:17:50-05:00

yindo commented

2026-02-16 09:17:50 -05:00

📋 Pull Request Information

Original PR: https://github.com/langchain-ai/openwork/pull/14
Author: @MohtashamMurshid
Created: 1/17/2026
Status: ✅ Merged
Merged: 1/17/2026
Merged by: @hntrl

Base: main ← Head: fix/pass-selected-model-to-backend

📝 Commits (1)

05e59aa fix: pass selected model to backend when invoking agent

📊 Changes

4 files changed (+29 additions, -20 deletions)

View changed files

📝 src/main/ipc/agent.ts (+9 -7)
📝 src/preload/index.ts (+7 -5)
📝 src/renderer/src/components/chat/ChatContainer.tsx (+6 -3)
📝 src/renderer/src/lib/electron-transport.ts (+7 -5)

📄 Description

Summary

Fixes the bug where selecting a Gemini (or OpenAI) model in the UI still resulted in "Anthropic API key not configured" error
The selected model was stored in UI state but never passed through the IPC chain to the backend

Changes

The fix passes modelId through the entire IPC chain:

ChatContainer.tsx - Include model_id in stream.submit() config
electron-transport.ts - Extract model_id from config and pass to streamAgent()
preload/index.ts - Forward modelId in IPC messages to main process
agent.ts - Extract modelId and pass to createAgentRuntime()

Test plan

Select a Gemini model in the model switcher
Configure only a Google API key (no Anthropic key)
Send a message
Verify the agent uses the Gemini model instead of erroring about missing Anthropic key

Fixes #12

_{🔄 This issue represents a GitHub Pull Request. It cannot be merged through Gitea due to API limitations.}

## 📋 Pull Request Information **Original PR:** https://github.com/langchain-ai/openwork/pull/14 **Author:** [@MohtashamMurshid](https://github.com/MohtashamMurshid) **Created:** 1/17/2026 **Status:** ✅ Merged **Merged:** 1/17/2026 **Merged by:** [@hntrl](https://github.com/hntrl) **Base:** `main` ← **Head:** `fix/pass-selected-model-to-backend` --- ### 📝 Commits (1) - [`05e59aa`](https://github.com/langchain-ai/openwork/commit/05e59aa1e089ec99ab782b900895669018b8db07) fix: pass selected model to backend when invoking agent ### 📊 Changes **4 files changed** (+29 additions, -20 deletions) <details> <summary>View changed files</summary> 📝 `src/main/ipc/agent.ts` (+9 -7) 📝 `src/preload/index.ts` (+7 -5) 📝 `src/renderer/src/components/chat/ChatContainer.tsx` (+6 -3) 📝 `src/renderer/src/lib/electron-transport.ts` (+7 -5) </details> ### 📄 Description ## Summary - Fixes the bug where selecting a Gemini (or OpenAI) model in the UI still resulted in "Anthropic API key not configured" error - The selected model was stored in UI state but never passed through the IPC chain to the backend ## Changes The fix passes `modelId` through the entire IPC chain: 1. **ChatContainer.tsx** - Include `model_id` in `stream.submit()` config 2. **electron-transport.ts** - Extract `model_id` from config and pass to `streamAgent()` 3. **preload/index.ts** - Forward `modelId` in IPC messages to main process 4. **agent.ts** - Extract `modelId` and pass to `createAgentRuntime()` ## Test plan - [x] Select a Gemini model in the model switcher - [x] Configure only a Google API key (no Anthropic key) - [x] Send a message - [x] Verify the agent uses the Gemini model instead of erroring about missing Anthropic key Fixes #12 --- <sub>🔄 This issue represents a GitHub Pull Request. It cannot be merged through Gitea due to API limitations.</sub>

yindo added the pull-request label 2026-02-16 09:17:50 -05:00

yindo closed this issue

2026-02-16 09:17:50 -05:00

Sign in to join this conversation.

1 Participants

Notifications

Due Date

No due date set.

Dependencies

No dependencies set.

Reference: langchain-ai/openwork#27