Bohdan Triapitsyn
28f0736d69
Adds a server-side "small model" capability: direct, cheap LLM calls that reuse the user's existing OpenCode provider logins — the mechanism OpenCode uses internally for titles and summaries but does not expose through the SDK or plugins. Zero new dependencies; plain fetch with per-provider wire formats, credentials never leave the server. Core (packages/web/server/lib/small-model): - Resolution mirrors OpenCode's session scoping: explicit settings override → small_model from the OpenCode config → family scan within the session's provider → the session's own model. The global provider scan only serves callers without a session context, and background callers forbid it entirely (restrictToPreferredProvider), so conversation content never reaches a provider the user didn't pick — explicit choices excepted. - Per-provider auth replicating OpenCode's plugin loaders: GitHub Copilot (device token as bearer, no exchange), ChatGPT plan via the codex Responses API (single-flight OAuth refresh written back to auth.json), Anthropic messages, Google generateContent, generic OpenAI-compatible. - OpenCode's free models (opencode/big-pickle, *-free) are never called directly; unauthenticated providers are skipped by design. - Prompt clamping to the model's catalog context limit; thinking disabled where a wire switch exists (Z.AI/GLM, MiniMax-M3, Gemini Flash); robust content parsing with a clear error when a thinking model spends its whol budget on reasoning. - Settings → Sessions gains a Small Model group: use-default checkbox plus an override picker limited to authenticated providers, persisted with web/desktop/VS Code sanitization parity. Consumers: - Session assist: a server-side watcher on the global SSE hub generates a short recap and one suggested follow-up after a session idles quietly fo a minute, stored on session metadata (openchamber.assist). Freshness is keyed to the last assistant message id, so new activity invalidates the payload everywhere with no extra writes. The chat shows the recap under the last message after five quiet minutes and the suggestion as a dismissible chip above the composer (tap fills the input, never sends). Gated by a new Chat setting (default on) that is a hard generation switch. Language is anchored to the conversation itself, with a script-mismatch guard against model/backend language hallucination. - TTS: a third input mode, summarized — long replies are condensed to spoken prose before playback on any TTS engine. - Git: commit-message and PR generation moved off the active chat session onto the small model fed with real diffs and the commit list (bodies included), with a session-transport fallback for free-model-only setups. - Notes: Add to notes distills long selections into 1-3 dense sentences preserving exact identifiers, with verbatim fallback on failure. Fixes along the way: - The global event watcher now starts unconditionally; it was gated behind the desktop-notify env, leaving the server-side event hub dead in packaged apps. - OpenCode re-emits message.updated for old user messages after idle; the watcher no longer mistakes those for new activity. - Session metadata merges from a fresh read right before the PATCH, so writes made during the generation window (suggestion dismissals, review links) are preserved; the assist runtime stops during graceful shutdown.
OpenChamber VS Code Extension
OpenCode AI coding agent, right inside your editor. No tab-switching, no context loss.
Like the extension? There's also a desktop app and web version with even more features.
What you get
- Chat beside your code — responsive layout that adapts to narrow and wide panels
- Agent Manager — run the same prompt across multiple models in parallel, compare results side by side
- Right-click actions — add context, explain selections, and improve code in-place
- Click-to-open — file paths in tool output open directly in your editor; edit-style results land in a focused diff view
- Session editor panel — keep chat sessions open alongside files
- Theme-aware — adapts to your VS Code light, dark, and high-contrast themes
Plus everything from the shared OpenChamber UI: branchable timeline, smart tool UIs, voice mode, Git workflows, and more.
Commands
| Command | Description |
|---|---|
OpenChamber: Focus Chat |
Focus the chat panel |
OpenChamber: New Session |
Start a new chat session |
OpenChamber: Open Sidebar |
Open the OpenChamber sidebar |
OpenChamber: Open Agent Manager |
Launch parallel multi-model runs |
OpenChamber: Open Session in Editor |
Open current or new session in an editor tab |
OpenChamber: Settings |
Open extension settings |
OpenChamber: Restart API Connection |
Restart the OpenCode API process |
OpenChamber: Show OpenCode Status |
Debug info for development or bug reports |
Right-click menu
Select code in the editor, right-click, and find the OpenChamber submenu:
| Action | Description |
|---|---|
| Add to Context | Attach selection to your next prompt |
| Explain | Ask the agent to explain the selected code |
| Improve Code | Ask the agent to improve the selection in-place |
Configuration
| Setting | Default | Description |
|---|---|---|
openchamber.apiUrl |
(empty) | URL of an external OpenCode API server. Leave empty to auto-start a local instance. |
openchamber.opencodeBinary |
(empty) | Absolute path to the opencode CLI binary. Useful when PATH lookup fails. Requires window reload to apply. |
Requirements
- OpenCode CLI installed and available in PATH (or set
OPENCODE_BINARYenv var) - VS Code 1.85+
Development
bun install
bun run vscode:dev
bun run vscode:dev now starts watchers + opens an Extension Development Host automatically. Webview UI changes use Vite HMR automatically.
Optional overrides:
OPENCHAMBER_VSCODE_BIN=cursor bun run vscode:devOPENCHAMBER_VSCODE_DEV_WORKSPACE=/path/to/workspace bun run vscode:devbun run vscode:dev /path/to/workspace
To package manually:
bun run --cwd packages/vscode build
cd packages/vscode && bunx vsce package --no-dependencies
Install locally: code --install-extension packages/vscode/openchamber-*.vsix
License
MIT