Pull Request Explorer
| Title | Author | Size | AI | Cycle Time | Review | Merged |
|---|---|---|---|---|---|---|
| feat(attio): add AI actions and output schemas | kishanprmr | XL | No | 0.3h | 0.1h | Oct 05, 2026 |
|
LLM analysis pending This PR has not been analyzed yet. |
||||||
| feat(chat): put a solution's flows and tables in one folder | hazemadelkhalel | L | No | 17.5h | 0.0h | Oct 05, 2026 |
|
LLM analysis pending This PR has not been analyzed yet. |
||||||
| feat(reddit): add AI actions and output schemas | kishanprmr | XL | No | 1.2h | 0.1h | Oct 05, 2026 |
|
LLM analysis pending This PR has not been analyzed yet. |
||||||
| feat(ai): run platform tier flow steps with ranked fallback models | Louai-Zokerburg | XL | AI | 95.9h | 0.1h | Oct 05, 2026 |
|
LLM analysis pending This PR has not been analyzed yet. |
||||||
| feat(cloudinary): add AI actions and output schemas | kishanprmr | XL | No | 24.8h | 0.1h | Oct 05, 2026 |
|
LLM analysis pending This PR has not been analyzed yet. |
||||||
| feat(presentation): add AI actions and output schemas | kishanprmr | XL | No | 67.6h | 0.1h | Oct 05, 2026 |
|
LLM analysis pending This PR has not been analyzed yet. |
||||||
| feat(supadata): add AI actions and output schemas | kishanprmr | XL | No | 69.1h | 0.1h | Oct 05, 2026 |
|
LLM analysis pending This PR has not been analyzed yet. |
||||||
| feat(grist): add AI actions and output schemas | kishanprmr | XL | No | 66.0h | 0.2h | Oct 05, 2026 |
|
LLM analysis pending This PR has not been analyzed yet. |
||||||
| feat(ticktick): add AI actions and output schemas | kishanprmr | XL | No | 64.9h | 0.1h | Oct 05, 2026 |
|
LLM analysis pending This PR has not been analyzed yet. |
||||||
| feat(zendesk): add AI actions and output schemas | kishanprmr | XL | No | 22.4h | 0.1h | Oct 05, 2026 |
|
LLM analysis pending This PR has not been analyzed yet. |
||||||
| feat(microsoft-todo): add AI actions and output schemas | kishanprmr | XL | No | 0.8h | 0.1h | Oct 05, 2026 |
|
LLM analysis pending This PR has not been analyzed yet. |
||||||
| fix(files): send upload file names as ASCII-safe headers | amrdb | M | AI | 120.1h | 0.1h | Oct 04, 2026 |
|
LLM analysis pending This PR has not been analyzed yet. |
||||||
| fix(helm): OpenShift overlay, and the cluster's default storage class when none is set | amrdb | M | AI | 77.0h | 0.1h | Oct 04, 2026 |
|
LLM analysis pending This PR has not been analyzed yet. |
||||||
| fix(connections): searching OAuth2 permissions filters the list again | majewskibartosz | L | No | 218.5h | 0.1h | Oct 04, 2026 |
|
LLM analysis pending This PR has not been analyzed yet. |
||||||
| feat(web): report which platform admin pages and buttons admins use, and where their plan stops them | AhmadTash | XL | No | 341.7h | 0.1h | Oct 04, 2026 |
|
LLM analysis pending This PR has not been analyzed yet. |
||||||
| fix(helm): use the cluster's default storage class when none is set | amrdb | S | AI | 0.1h | 0.1h | Oct 04, 2026 |
|
LLM analysis pending This PR has not been analyzed yet. |
||||||
| fix(connections): replacing a connection keeps draft edits and leaves disabled flows disabled | majewskibartosz | L | No | 213.1h | 0.1h | Oct 04, 2026 |
|
LLM analysis pending This PR has not been analyzed yet. |
||||||
| fix(chat): long chats no longer fail a turn while older messages are being summarized | hazemadelkhalel | M | AI | 4.6h | 0.0h | Oct 04, 2026 |
|
LLM analysis pending This PR has not been analyzed yet. |
||||||
| test(agent-eval): publish nightly eval runs and measure the judge against humans | Louai-Zokerburg | XL | AI | 27.1h | 0.1h | Oct 04, 2026 |
|
LLM analysis pending This PR has not been analyzed yet. |
||||||
| test(agent-eval): make chat eval verdicts trustworthy | Louai-Zokerburg | M | AI | 27.5h | 0.1h | Oct 04, 2026 |
|
LLM analysis pending This PR has not been analyzed yet. |
||||||
| feat(web): redesign the MCP admin pages around four tool tiers | moaaz-ae | XL | No | 122.2h | 0.1h | Oct 04, 2026 |
|
LLM analysis pending This PR has not been analyzed yet. |
||||||
| fix(agents): an outage at your own AI provider no longer shows up as a platform failure | hazemadelkhalel | S | AI | 0.3h | 0.1h | Oct 04, 2026 |
|
LLM analysis pending This PR has not been analyzed yet. |
||||||
| refactor(web): colour is numbered steps, and every call site moved onto it | moaaz-ae | XL | No | 169.6h | 0.1h | Oct 04, 2026 |
|
LLM analysis pending This PR has not been analyzed yet. |
||||||
| feat(ai): platform model tiers data and api | Louai-Zokerburg | XL | AI | 85.6h | 0.1h | Oct 04, 2026 |
|
LLM analysis pending This PR has not been analyzed yet. |
||||||
| fix(web): tighten the Roles & Access copy | lina-alomari | M | No | 0.3h | 0.0h | Oct 04, 2026 |
|
LLM analysis pending This PR has not been analyzed yet. |
||||||
| feat(google-my-business): ui improvement | Talaljaber | XL | No | 5.6h | 0.1h | Oct 03, 2026 |
|
LLM analysis pending This PR has not been analyzed yet. |
||||||
| feat(stripe): ui improvement | Talaljaber | XL | No | 5.1h | 0.1h | Oct 03, 2026 |
|
LLM analysis pending This PR has not been analyzed yet. |
||||||
| feat(apitable): ui improvement | Talaljaber | XL | No | 5.7h | 0.1h | Oct 03, 2026 |
|
LLM analysis pending This PR has not been analyzed yet. |
||||||
| feat(perplexity-ai): ui improvement | Talaljaber | L | No | 5.9h | 0.1h | Oct 03, 2026 |
|
LLM analysis pending This PR has not been analyzed yet. |
||||||
| feat(linear): ui improvement | Talaljaber | XL | No | 30.6h | 0.1h | Oct 03, 2026 |
|
LLM analysis pending This PR has not been analyzed yet. |
||||||
| feat(ai): ui improvement | Talaljaber | M | No | 52.0h | 0.1h | Oct 03, 2026 |
|
LLM analysis pending This PR has not been analyzed yet. |
||||||
| feat(claude): ui improvement | Talaljaber | L | AI | 52.3h | 0.1h | Oct 03, 2026 |
|
LLM analysis pending This PR has not been analyzed yet. |
||||||
| feat(rss): ui improvement | Talaljaber | M | No | 52.6h | 0.2h | Oct 03, 2026 |
|
LLM analysis pending This PR has not been analyzed yet. |
||||||
| feat(google-gemini): clearer forms, aspect ratio cards and a shorter Create Video step | Talaljaber | M | AI | 166.8h | 0.1h | Oct 03, 2026 |
|
LLM analysis pending This PR has not been analyzed yet. |
||||||
| feat(imap): clearer connection setup, full folder paths and a delete warning | Talaljaber | M | No | 171.3h | 0.1h | Oct 03, 2026 |
|
LLM analysis pending This PR has not been analyzed yet. |
||||||
| feat(hubspot): readable property notes, plain-language labels and optional fields under Advanced | Talaljaber | XL | No | 188.7h | 0.1h | Oct 03, 2026 |
|
LLM analysis pending This PR has not been analyzed yet. |
||||||
| feat(csv): plain-language step forms, and JSON to CSV keeps every column | Talaljaber | M | No | 171.3h | 0.1h | Oct 03, 2026 |
|
LLM analysis pending This PR has not been analyzed yet. |
||||||
| feat(youtube): shorter Search form, tabbed channel and playlist lookups, and plain field help | Talaljaber | L | No | 189.8h | 0.1h | Oct 03, 2026 |
|
LLM analysis pending This PR has not been analyzed yet. |
||||||
| feat(groq): ui improvement | Talaljaber | L | AI | 29.8h | 0.1h | Oct 03, 2026 |
|
LLM analysis pending This PR has not been analyzed yet. |
||||||
| feat(twilio): ui improvement | Talaljaber | L | No | 51.2h | 0.2h | Oct 03, 2026 |
|
LLM analysis pending This PR has not been analyzed yet. |
||||||
| feat(twitter): ui improvement | Talaljaber | L | No | 29.5h | 0.1h | Oct 03, 2026 |
|
LLM analysis pending This PR has not been analyzed yet. |
||||||
| feat(openai): ui improvement | Talaljaber | XL | AI | 29.5h | 0.1h | Oct 03, 2026 |
|
LLM analysis pending This PR has not been analyzed yet. |
||||||
| fix(worker): record run status and engine error on the job.execute event | amrdb | M | AI | 49.6h | 0.1h | Oct 01, 2026 |
SummaryBug FixRecord run status and engine error on job.execute Improves error visibility for self-hosted customers by embedding detailed run status and error information in job.execute events, enabling faster debugging and alerting. Health Assessment
Small
Low
Low
AI Details
Usage:
Authored by AI
Category:
Code AI
Tools:
Claude
Confidence:
0.95
Tech Stack
Languages:
Typescript
|
||||||
| feat(agents): the runs and configure panels open and close smoothly, with buttons that stay in place | hazemadelkhalel | L | No | 0.8h | 0.1h | Oct 01, 2026 |
SummaryFeatureSmooth agent panel transitions with persistent buttons Improves agent page usability by providing smooth side panel transitions and persistent action buttons, reducing user confusion and enhancing workflow efficiency. Users can now switch between runs and configure panels without layout shifts, improving productivity. Health Assessment
Large
Medium
Low
AI Details
Confidence:
0.20
Tech Stack
Languages:
Typescript
Frameworks:
React, Nextjs
|
||||||
| feat(flowlu): manage CRM accounts, opportunities, tasks and projects, with new-record triggers | OdaiAhmed99 | XL | AI | 5.7h | 0.1h | Oct 01, 2026 |
SummaryFeatureAdd Flowlu CRM integration with new-record triggers Enables users to manage Flowlu CRM entities and automate workflows with new triggers, improving productivity and data consistency. Health Assessment
X-Large
Low
Low
AI Details
Usage:
Authored by AI
Category:
Both AI
Tools:
Claude, Greptile
Confidence:
0.95
Tech Stack
Languages:
Typescript
|
||||||
| feat(google-slides): build and edit slides, text, shapes, tables and images, and export decks | OdaiAhmed99 | XL | AI | 8.0h | 0.1h | Oct 01, 2026 |
SummaryFeatureAdd Google Slides integration for building and editing decks Enables users to create, edit, and export Google Slides decks directly within Activepieces, expanding automation capabilities and streamlining presentation workflows. Health Assessment
X-Large
Low
Low
AI Details
Usage:
Authored by AI
Category:
Code AI
Tools:
Claude
Confidence:
0.95
Tech Stack
Languages:
Typescript
|
||||||
| feat(binance): add prices, 24h stats, order book, trades, candles and trading pairs | OdaiAhmed99 | XL | AI | 8.0h | 0.1h | Oct 01, 2026 |
SummaryFeatureAdd Binance market data actions Introduces ten new read‑only actions for Binance, enabling agents to retrieve prices, order books, trades, candles, and trading pairs without a connection, enhancing data‑driven workflows. Health Assessment
X-Large
Low
Low
AI Details
Usage:
Authored by AI
Category:
Code AI
Tools:
Claude
Confidence:
0.95
Tech Stack
Languages:
Typescript
|
||||||
| feat(agents): an agent's runs open in a side panel next to the chat, like the flow builder | hazemadelkhalel | L | No | 0.6h | 0.1h | Oct 01, 2026 |
SummaryFeatureAdd agent runs side panel next to chat Users can now view and manage agent runs directly in a side panel, improving workflow visibility and reducing context switching. Health Assessment
Large
Medium
Low
AI Details
Confidence:
0.95
Tech Stack
Languages:
Typescript
Frameworks:
React, Nextjs
|
||||||
| feat(chat): chat on cloud follows the plan instead of a 200-user beta cap | hazemadelkhalel | L | No | 3.3h | 0.1h | Oct 01, 2026 |
SummaryFeatureChat on Cloud now plan-controlled, no beta cap Enables chat for all Cloud users based on plan settings, removing the 200-user beta cap and simplifying rollout. Health Assessment
Large
Medium
Low
AI Details
Confidence:
0.95
Tech Stack
Languages:
Typescript, Javascript
|
||||||
| fix(ai): stop offering Google models that new Google projects cannot use | hazemadelkhalel | M | No | 1.8h | 0.1h | Oct 01, 2026 |
SummaryBug FixFix AI model picker for new Google projects Ensures new Google projects don't encounter deprecated Gemini models, improving reliability and user experience. Health Assessment
Medium
Low
Low
AI DetailsTech Stack
Languages:
Typescript, Javascript
|
||||||