Pull Request Explorer
| Title | Author | Size | AI | Cycle Time | Review | Merged |
|---|---|---|---|---|---|---|
| perf: batch and deduplicate per-request DB reads in the chat middleware | Classic298 | M | No | 76.6h | - | Jul 23, 2026 |
|
LLM analysis pending This PR has not been analyzed yet. |
||||||
| feat: add LDAP group synchronization support | gaby | M | No | 61.0h | 30.1h | Jul 23, 2026 |
|
LLM analysis pending This PR has not been analyzed yet. |
||||||
| fix: correct async import hook that disables the code-interpreter module blocklist | Classic298 | XS | No | 67.2h | - | Jul 23, 2026 |
|
LLM analysis pending This PR has not been analyzed yet. |
||||||
| feat: forward user info headers to Mistral OCR API | andrep2222 | S | AI | 65.5h | 0.0h | Jul 23, 2026 |
|
LLM analysis pending This PR has not been analyzed yet. |
||||||
| perf: deduplicate repeated config fetches in Ollama request handlers | Classic298 | M | No | 75.0h | - | Jul 23, 2026 |
|
LLM analysis pending This PR has not been analyzed yet. |
||||||
| perf: stop recomputing static work in per-response ASGI middlewares | Classic298 | S | No | 74.6h | - | Jul 23, 2026 |
|
LLM analysis pending This PR has not been analyzed yet. |
||||||
| perf: linear content accumulation in the streaming response handler | Classic298 | S | No | 74.5h | - | Jul 23, 2026 |
|
LLM analysis pending This PR has not been analyzed yet. |
||||||
| fix: only admins may switch the global Automatic1111 checkpoint from image generation | Classic298 | XS | No | 67.2h | - | Jul 23, 2026 |
|
LLM analysis pending This PR has not been analyzed yet. |
||||||
| fix: enforce action availability and model access on the chat action route | Classic298 | S | No | 67.4h | - | Jul 23, 2026 |
|
LLM analysis pending This PR has not been analyzed yet. |
||||||
| perf: halve function-table queries in get_all_models | Classic298 | S | No | 74.3h | - | Jul 23, 2026 |
|
LLM analysis pending This PR has not been analyzed yet. |
||||||
| perf: 40% LESS CPU usage: cut per-instance CPU cost of shared socket.io Redis pub/sub channel | Classic298 | S | No | 51.6h | 0.6h | Jul 23, 2026 |
|
LLM analysis pending This PR has not been analyzed yet. |
||||||
| fix: Log upstream provider rejections at warn/error level | Classic298 | S | No | 63.7h | - | Jul 23, 2026 |
|
LLM analysis pending This PR has not been analyzed yet. |
||||||
| fix: add missing folders field to SharingPermissions so the permission persists | Classic298 | XS | No | 38.0h | - | Jul 23, 2026 |
|
LLM analysis pending This PR has not been analyzed yet. |
||||||
| **i18n**: Internationalization or localization changes | andibing | S | AI | 33.9h | - | Jul 23, 2026 |
|
LLM analysis pending This PR has not been analyzed yet. |
||||||
| i18n: add pt-BR translations for newly added UI items and consistency pass | joaoback | XL | No | 17.0h | - | Jul 23, 2026 |
|
LLM analysis pending This PR has not been analyzed yet. |
||||||
| i18n: Update Japanese language translations | A-PS1999 | L | AI | 69.7h | - | Jul 10, 2026 |
SummaryChoreUpdate Japanese translations Adds missing Japanese translations and corrects errors, improving user experience for Japanese-speaking users. Health Assessment
Medium
Low
Low
AI Details
Category:
Code AI
Confidence:
0.80
Tech Stack |
||||||
| fix: don't mutate caller's payload when resolving a pipe base model | Classic298 | XS | No | 51.1h | - | Jul 10, 2026 |
SummaryBug FixFix payload mutation in pipe base model resolution Prevents unauthorized access errors after tool calls by keeping the original user payload unchanged, improving reliability for non‑admin users. Health Assessment
Small
Low
Medium
AI DetailsTech Stack
Languages:
Python
|
||||||
| fix: socket reconnect recovery never runs for chats started from the home page | Classic298 | XS | No | 44.3h | - | Jul 10, 2026 |
SummaryBug FixFix socket reconnect for home page chats Ensures chat sessions started from the home page recover from websocket interruptions, preventing users from getting stuck in a loading state. This improves user experience and reduces support tickets. Health Assessment
Small
Low
Low
AI DetailsTech Stack
Languages:
Javascript
Frameworks:
Svelte
|
||||||
| fix: gate allow-same-origin on the terminal file-preview iframe to prevent same-origin XSS | Classic298 | XS | No | 47.2h | - | Jul 10, 2026 |
SummaryBug FixFix XSS in terminal file preview iframe This change removes the allow-same-origin attribute from the terminal file preview iframe, preventing scripts in previewed HTML from accessing the application's origin. It mitigates potential account takeover and server-side RCE vulnerabilities. Health Assessment
Small
Low
Low
AI DetailsTech Stack
Frameworks:
Svelte
|
||||||
| fix: Milvus multitenancy scalar index creation on Milvus Lite | Classic298 | S | No | 44.6h | - | Jul 10, 2026 |
SummaryBug FixFix Milvus multitenancy scalar index creation Fixes embedding write failures in Milvus Lite when multitenancy mode is enabled, ensuring reliable memory add, file upload, and knowledge writes. Health Assessment
Small
Low
Low
AI Details
Confidence:
0.20
Tech Stack
Languages:
Python
|
||||||
| fix: convert SecurityHeadersMiddleware to pure ASGI | Classic298 | S | No | 35.1h | - | Jul 10, 2026 |
SummaryBug FixConvert SecurityHeadersMiddleware to pure ASGI Improves reliability of audio streaming responses by converting middleware to pure ASGI, preventing buffering issues. Health Assessment
Small
Low
Low
AI Details
Confidence:
1.00
Tech Stack
Languages:
Python
|
||||||
| fix: restore prompt confirmation for sandboxed tool result embeds | Classic298 | S | No | 44.0h | - | Jul 10, 2026 |
SummaryBug FixRestore prompt confirmation for sandboxed tool embeds Ensures that prompts from sandboxed tool embeds trigger a confirmation dialog, preventing unintended submissions and enhancing user safety. This restores the intended Rich UI flow and protects against accidental prompt execution. Health Assessment
Small
Low
Low
AI Details
Confidence:
1.00
Tech Stack |
||||||
| 0.10.2 | tjbck | XL | No | 0.0h | - | Jul 01, 2026 |
SummaryHealth AssessmentAI DetailsTech Stack |
||||||
| Update CHANGELOG.md | Classic298 | S | No | 0.1h | - | Jul 01, 2026 |
SummaryDocsUpdate CHANGELOG.md Updates the project changelog to reflect recent changes, improving release transparency. Health Assessment
Small
Low
Low
AI Details
Confidence:
0.80
Tech Stack |
||||||
| fix: derive content from output for search | Algorithm5838 | XS | AI | 27.0h | - | Jul 01, 2026 |
SummaryBug FixFix: derive content from output for search Ensures messages generated via the output path are searchable by setting content appropriately, improving search functionality without affecting display. Health Assessment
Small
Low
Low
AI Details
Category:
Code AI
Confidence:
0.95
Tech Stack
Frameworks:
Svelte
|
||||||
| i18n: Update and fix Catalan translation.json | aleixdorca | L | No | 24.7h | - | Jul 01, 2026 |
SummaryBug FixUpdate Catalan translation file Improves Catalan language support by correcting translation errors, enhancing user experience for Catalan-speaking users. Health Assessment
Large
Low
Low
AI Details
Confidence:
0.95
Tech Stack |
||||||
| i18n: fix Spanish relative time labels | torgus | S | AI | 10.7h | - | Jul 01, 2026 |
SummaryBug FixFix Spanish relative time labels Improves Spanish UI clarity by correcting time labels, reducing confusion for Spanish-speaking users. Health Assessment
Small
Low
Low
AI Details
Category:
Code AI
Confidence:
0.80
Tech Stack |
||||||
| fix: use absolute indexURL for pyodide sandbox | Algorithm5838 | M | AI | 1.9h | - | Jul 01, 2026 |
SummaryBug FixFix absolute indexURL for Pyodide sandbox Ensures Pyodide loads correctly in the sandbox iframe, preventing execution timeouts and improving reliability for Python code execution. Health Assessment
Small
Low
Low
AI Details
Category:
Code AI
Confidence:
0.20
Tech Stack
Languages:
Typescript
|
||||||
| fix: use updated_at for sidebar chat timestamp | Classic298 | XS | No | 15.3h | - | Jul 01, 2026 |
SummaryBug FixFix sidebar chat timestamp to use updated_at Ensures sidebar timestamps reflect recent activity, preventing confusing stale labels and improving user experience. Health Assessment
Small
Low
Low
AI DetailsTech Stack |
||||||
| i18n: translate missing Thai keys | SNaytiP | L | AI | 14.7h | - | Jun 30, 2026 |
SummaryFeatureTranslate missing Thai localization strings Adds missing Thai translations, improving user experience for Thai-speaking users. Health Assessment
Large
Low
Low
AI Details
Category:
Code AI
Confidence:
0.20
Tech Stack |
||||||
| i18n: add pt-BR translations for newly added UI items and consistency pass | joaoback | L | No | 5.8h | - | Jun 30, 2026 |
SummaryFeatureAdd pt-BR translations for new UI items Adds Brazilian Portuguese translations for newly added UI elements and performs a consistency and quality pass on existing strings, improving clarity and user experience for pt-BR users. Health Assessment
Medium
Low
Low
AI Details
Confidence:
0.20
Tech Stack |
||||||
| 0.10.1 | tjbck | S | No | - | - | Jun 29, 2026 |
SummaryHealth AssessmentAI DetailsTech Stack |
||||||
| 0.10.0 | tjbck | XL | No | 1.3h | 0.1h | Jun 29, 2026 |
SummaryHealth AssessmentAI DetailsTech Stack |
||||||
| Update translation.json | Classic298 | S | No | 0.0h | - | Jun 29, 2026 |
SummaryChoreUpdate German translation file Adds updated German translations for UI, improving user experience for German-speaking users. Health Assessment
Small
Low
Low
AI Details
Confidence:
0.95
Tech Stack |
||||||
| perf: non-blocking model and tool server loading for unreachable connections | silentoplayz | S | No | 120.4h | - | Jun 29, 2026 |
SummaryBug FixImprove UI responsiveness for unreachable connections This PR removes UI lockup caused by slow or unreachable external services by loading models and tool servers asynchronously and skipping redundant backend calls, improving user experience and reducing page load times. Health Assessment
Small
Low
High
AI Details
Usage:
AI Assisted
Category:
Code AI
Tools:
Copilot
Confidence:
0.95
Tech Stack
Languages:
Python
Frameworks:
Fastapi
|
||||||
| fix: workspace models bulk actions and search pagination | silentoplayz | M | No | 130.1h | - | Jun 29, 2026 |
SummaryBug FixFix workspace models bulk actions and search pagination This fix ensures bulk actions and search on the Workspace Models page correctly affect all models across all pages, improving user productivity and preventing incomplete operations. Health Assessment
Medium
Low
High
AI Details
Usage:
AI Assisted
Category:
Code AI
Tools:
Copilot
Confidence:
0.95
Tech Stack |
||||||
| chore(deps): patch frontend security dependencies | gateway | M | No | 122.6h | - | Jun 29, 2026 |
SummaryChorePatch frontend security dependencies The PR patches several critical frontend dependencies to remove known security vulnerabilities, ensuring the application remains secure. No user-facing changes are expected. Health Assessment
Medium
Medium
Medium
AI Details
Confidence:
0.95
Tech Stack |
||||||
| fix: pinned note sidebar drag-and-drop into chat context | silentoplayz | S | AI | 0.0h | - | Jun 29, 2026 |
SummaryBug FixFix pinned note drag-and-drop into chat Pinned notes can now be dragged from the sidebar into the chat area to attach as context references, restoring expected drag-and-drop functionality. This resolves silent failures reported in #25771. Health Assessment
Small
Low
Low
AI Details
Usage:
AI Assisted
Category:
Code AI
Tools:
Copilot
Confidence:
0.95
Tech Stack
Frameworks:
Svelte
|
||||||
| i18n: complete German (de-DE) translations | Classic298 | L | No | 0.0h | - | Jun 29, 2026 |
SummaryFeaturei18n: complete German (de-DE) translations Completes German language support for the web UI, enhancing accessibility for German-speaking users. Health Assessment
Small
Low
Low
AI Details
Confidence:
0.75
Tech Stack |
||||||
| fix: allow special tokens in document text during chunk measurement | silentoplayz | XS | No | 228.1h | - | Jun 29, 2026 |
SummaryBug FixFix special token handling in chunk measurement This fix stops errors when documents contain LLM special token strings, allowing them to be indexed correctly. It ensures reliable document ingestion for knowledge bases. Health Assessment
Small
Low
High
AI Details
Usage:
AI Assisted
Category:
Code AI
Tools:
Copilot
Confidence:
0.95
Tech Stack
Languages:
Python
|
||||||
| perf(backend): offload blocking calls in async paths to threads | gaby | S | AI | 0.8h | - | Jun 29, 2026 |
SummaryRefactorOffload blocking calls to threads for async performance Improves backend async performance by moving blocking operations to background threads, reducing event loop latency and enhancing responsiveness. This change ensures smoother user experience during heavy I/O tasks. Health Assessment
Small
Low
Low
AI Details
Category:
Code AI
Confidence:
0.75
Tech Stack
Languages:
Python
|
||||||
| feat: forward client User-Agent to ALL model backends | Classic298 | S | No | 68.7h | 0.7h | Jun 29, 2026 |
SummaryFeatureForward client User-Agent to all model backends Enables upstream model providers to see the real client User-Agent, improving usage attribution, cost tracking, and telemetry. Health Assessment
Small
Low
Low
AI Details
Confidence:
0.10
Tech Stack
Languages:
Python
Frameworks:
Svelte
|
||||||
| perf: make ToolCallDisplay parseJSONString iterative | Classic298 | S | No | 259.9h | - | Jun 29, 2026 |
SummaryRefactorImprove ToolCallDisplay JSON parsing performance Optimizes JSON parsing in ToolCallDisplay to avoid recursion stack overflows, improving performance for tool-call payloads. This change reduces CPU usage and eliminates silent failures during parsing. Health Assessment
Small
Low
High
AI DetailsTech Stack
Languages:
Javascript
Frameworks:
Svelte
|
||||||
| fix: prevent Enter key symbol rendering as emoji on iOS in Keyboard Shortcuts modal | silentoplayz | XS | No | 246.6h | - | Jun 29, 2026 |
SummaryBug FixFix iOS emoji rendering in Keyboard Shortcuts Corrects a visual bug where the Enter key symbol appears as an emoji on iOS, ensuring consistent text glyphs and aligning menu capitalization across platforms. Health Assessment
Small
Low
Medium
AI Details
Usage:
AI Assisted
Category:
Code AI
Tools:
Copilot
Confidence:
0.95
Tech Stack
Languages:
Javascript
Frameworks:
Svelte
|
||||||
| fix: strip whitespace from user info headers to prevent MCP connection failures | HrushiYadav | XS | AI | 237.0h | - | Jun 29, 2026 |
SummaryBug FixFix whitespace in user headers to prevent MCP failures This fix prevents connection failures for users whose display names contain leading or trailing whitespace. It ensures reliable communication with MCP servers, improving user experience. Health Assessment
Small
Low
Low
AI Details
Category:
Code AI
Confidence:
0.95
Tech Stack
Languages:
Python
|
||||||
| fix: solve-valves-icon-disappear-issue | alvarellos | XS | AI | 142.2h | - | Jun 29, 2026 |
SummaryBug FixFix valve icon disappearance in UI Resolves intermittent disappearance of tool valve icons, ensuring consistent UI visibility for users. Improves reliability of the tool integration menu. Health Assessment
Small
Low
Low
AI Details
Category:
Code AI
Confidence:
0.20
Tech Stack
Languages:
Python
|
||||||
| perf: skip re-sanitizing unchanged HTML tokens | Algorithm5838 | XS | AI | 74.9h | - | Jun 29, 2026 |
SummaryRefactorPerformance: Skip re-sanitizing unchanged HTML tokens Improves rendering performance for streaming messages by avoiding redundant sanitization, ensuring faster UI updates without altering user experience. Health Assessment
Small
Low
Low
AI Details
Category:
Code AI
Confidence:
0.75
Tech Stack
Frameworks:
Svelte
|
||||||
| fix: ignore IME composition Enter in chat search (#26172) | ybnjm1234-source | S | AI | 161.2h | - | Jun 29, 2026 |
SummaryBug FixFix IME composition Enter in chat search Prevents accidental search actions when confirming IME composition in chat search, improving user experience for East Asian language users. Health Assessment
Small
Low
Low
AI Details
Category:
Code AI
Confidence:
0.20
Tech Stack |
||||||
| i18n: add pt-BR translations for newly added UI items and consistency pass | joaoback | M | No | 80.0h | - | Jun 29, 2026 |
SummaryFeatureAdd pt-BR translations for new UI items Adds Brazilian Portuguese translations for newly added UI elements and improves consistency across existing strings, enhancing clarity and user experience for pt-BR users. Health Assessment
Medium
Low
Medium
AI Details
Confidence:
0.20
Tech Stack |
||||||
| fix: persist control edits across navigation | Algorithm5838 | XS | AI | 46.7h | - | Jun 29, 2026 |
SummaryBug FixPersist chat control edits across navigation Ensures that changes to chat controls, such as system prompts and attached files, are retained when switching between chats, improving user experience. Health Assessment
Small
Low
Low
AI Details
Category:
Code AI
Confidence:
0.95
Tech Stack |
||||||