[GH-ISSUE #1366] Feature Request: Add persistent custom instructions / memory across conversations #905

New issue

Open

opened 2026-10-26 21:32:02 +00:00 by bsakaguchi · 2 comments

bsakaguchi commented

2026-10-26 21:32:02 +00:00

Owner

Originally created by @Gitcoder12 on GitHub (May 28, 2026).
Original GitHub issue: https://github.com/deepseek-ai/DeepSeek-V3/issues/1366

Currently DeepSeek has no way to retain user preferences, prompts, or instructions across different chat sessions. This limits its usefulness for users who want consistent behavior (e.g., "never show thinking process", "always answer concisely", "highlight certain topics").

Request:

Add a "Memory" or "Custom Instructions" section in settings (like ChatGPT or Claude)
Allow users to save a prompt that gets injected into every new conversation
(Optional) Let users edit or disable it at any time

Use case examples:

Exam prep: user wants step-by-step reasoning without meta-commentary
Work projects: user wants structured, long-form answers every time
Personal preference: user doesn't want to see internal thinking tags

This is a highly requested feature. Please consider implementing it.

Related issues: #1106, #1190

Originally created by @Gitcoder12 on GitHub (May 28, 2026). Original GitHub issue: https://github.com/deepseek-ai/DeepSeek-V3/issues/1366 Currently DeepSeek has no way to retain user preferences, prompts, or instructions across different chat sessions. This limits its usefulness for users who want consistent behavior (e.g., "never show thinking process", "always answer concisely", "highlight certain topics"). Request: - Add a "Memory" or "Custom Instructions" section in settings (like ChatGPT or Claude) - Allow users to save a prompt that gets injected into every new conversation - (Optional) Let users edit or disable it at any time Use case examples: - Exam prep: user wants step-by-step reasoning without meta-commentary - Work projects: user wants structured, long-form answers every time - Personal preference: user doesn't want to see internal thinking tags This is a highly requested feature. Please consider implementing it. Related issues: #1106, #1190

bsakaguchi commented

2026-10-26 21:32:03 +00:00

Author

Owner

@Equinox984 commented on GitHub (Jun 17, 2026):

Yep, I think this is one of the most important features an AI should have imo

@Equinox984 commented on GitHub (Jun 17, 2026): Yep, I think this is one of the most important features an AI should have imo

bsakaguchi commented

2026-10-26 21:32:03 +00:00

Author

Owner

@github-actions[bot] commented on GitHub (Jul 31, 2026):

This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs. If you believe this issue is still relevant, please leave a comment to keep it open. Thank you for your contributions!

@github-actions[bot] commented on GitHub (Jul 31, 2026): This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs. If you believe this issue is still relevant, please leave a comment to keep it open. Thank you for your contributions!