Title: File Context is Not Shared Sequentially Across All Models in Multi-Agent Mode ​Body: I’ve noticed a critical issue with context/file sharing when running multiple agents simultaneously. ​When I uploaded a PDF manuscript (ASCOM-D-26-00177.pdf) and asked for an analysis, only a few models (like ChatGPT 5.5 and Gemini 3.1 Pro) successfully recognized and summarized the document. ​However, at the exact same time, other models (like Claude Sonnet 4.6 and Claude Opus 4.7) completely failed to read the uploaded file. Instead of analyzing it, they just outputted their default greeting templates (e.g., "What would you like me to analyze? Please paste your text..."). ​The core strength of the Multi-Agent mode should be "cross-validation"—allowing users to compare how different models analyze the exact same data to spot overlapping issues. Currently, because the file context isn't being properly fed to all selected models simultaneously, this collective intelligence feature is broken. ​Please review the multi-age

Feedback on Deep Search Mode: Answers are less detailed and shorter than when turned off

​[Issue Description]

When asking broad conceptual questions such as "What is the universe?" (우주란), the standard mode (Deep Search OFF) provides a rich, friendly, and detailed response filled with specific examples like stars, planets, galaxies, and gravity, along with scientific and philosophical contexts. (Please refer to attached files 1000005693.jpg and 1000005695.jpg).

​However, when Deep Search is turned ON, the response starts with a very dry, single-line dictionary definition ("The universe is the sum of everything that exists") and is significantly shorter and more overly compressed than the standard mode. (Please refer to attached files 1000005697.jpg and 1000005699.jpg).

​[Suggestions for Improvement]

Users naturally expect "Deep Search" to deliver much deeper, more thorough analysis and advanced information (e.g., modern cosmological theories, expanded data, etc.) compared to the standard mode. Currently, the system seems to interpret "Deep Search" as "concise summarization," creating a gap between user expectations and the actual output.

​Even when a user inputs a single broad keyword, the Deep Search mode should expand upon the core definition by providing advanced context or academic background, rather than just spitting out a brief, over-summarized response. I hope the prompt interpretation logic can be improved to reflect this.

​Option 2: Brief & Direct (간결한 요약형)

​Subject: Deep Search mode provides less detailed answers than when it's off

​When I ask the same question ("What is the universe?"), the standard mode with Deep Search OFF (1000005693.jpg) actually gives a much more detailed explanation with plenty of examples.

​On the other hand, when I turn Deep Search ON (1000005697.jpg), the answer becomes way too simple and over-summarized, merely stating that it is "the sum of everything that exists."

​As a user, I expect more depth and details when turning on "Deep Search," so it feels counterintuitive that the content actually shrinks. Even for single-word queries, please improve the Deep Search mode so that it expands into deeper, more co

mprehensive information.