Commit graph

  • 1a83bbcc94 Clean API chat router. Move FeedbackData response type to router helper Debanjum 2024-11-01 17:31:43 -0700
  • e6eb87bbb5 Merge branch 'improve-debug-reasoning-and-other-misc-fixes' of github.com:khoj-ai/khoj into improve-debug-reasoning-and-other-misc-fixes sabaimran 2024-11-01 16:48:39 -0700
  • a213b593e8 Limit the number of urls the webscraper can extract for scraping sabaimran 2024-11-01 16:48:36 -0700
  • 327fcb8f62 create defiltered query after conversation command is extracted sabaimran 2024-11-01 16:48:03 -0700
  • b79a9ec36d Clarify description of the code evaluation environment: not for document creation sabaimran 2024-11-01 16:47:27 -0700
  • 9c7b36dc69 Use standard per minute rate limits across user types Debanjum 2024-11-01 16:15:09 -0700
  • ac21b10dd5 Simplify logic to get default search model. Remove unused import Debanjum 2024-11-01 15:14:00 -0700
  • 2b35790165 Merge branch 'master' of github.com:khoj-ai/khoj into improve-debug-reasoning-and-other-misc-fixes sabaimran 2024-11-01 14:51:26 -0700
  • 22f3ed3f5d
    Research Mode: Give Khoj the ability to perform more advanced reasoning (#952) Debanjum 2024-11-01 14:46:29 -0700
  • baa939f4ce When running code, strip any code delimiters. Disable application json type specification in Gemini request. sabaimran 2024-11-01 13:47:39 -0700
  • 8fd2fe162f Determine if research mode is enabled by checking the conversation commands and 'linting' them in the selection phase sabaimran 2024-11-01 13:12:34 -0700
  • cead1598b9 Don't reset research mode after completing research execution sabaimran 2024-11-01 13:00:11 -0700
  • c1c779a7ef Do not yaml format raw code results in context for LLM. It's confusing Debanjum 2024-11-01 12:04:38 -0700
  • b3dad1f393 Standardize rate limits to 1/6 ratio sabaimran 2024-11-01 12:21:09 -0700
  • 23a49b6b95 Add documentation for python code execution capability sabaimran 2024-11-01 12:14:33 -0700
  • cd75151431 Do not allow auto selecting research mode as tool for now. Debanjum 2024-11-01 11:58:44 -0700
  • 0b0cfb35e6 Simplify in research mode check in api_chat. Debanjum 2024-11-01 11:55:13 -0700
  • ffa7f95559 Add template for a code sandbox to the docker-compose configuration sabaimran 2024-11-01 11:50:58 -0700
  • 73750ef286 Merge branch 'master' into features/advanced-reasoning Debanjum 2024-11-01 11:42:01 -0700
  • 1fc280db35 Handle case where infer_webpage_url returns no valid urls sabaimran 2024-11-01 11:41:32 -0700
  • 1c920273dd
    Add Prompt Tracer to Visualize, Analyze and Debug Khoj's Train of Thought (#951) Debanjum 2024-11-01 11:33:54 -0700
  • 33d36ee58c Add experimental notice to research mode tooltip sabaimran 2024-11-01 11:00:27 -0700
  • 0145b2a366 Set usage limits on the research mode sabaimran 2024-11-01 10:29:33 -0700
  • 3ea94ac972 Only include inferred-queries in chat history when present sabaimran 2024-10-31 22:01:41 -0700
  • 149cbe1019 Use bottom anchor for the commandbar popover sabaimran 2024-10-31 20:40:38 -0700
  • 21858acccc Remove conversation command always in query, filter out inferred queries that were not with selected tool when going through tool selection iterations sabaimran 2024-10-31 20:27:38 -0700
  • 19241805ee Merge branch 'master' of github.com:khoj-ai/khoj into improve-debug-reasoning-and-other-misc-fixes sabaimran 2024-10-31 18:20:23 -0700
  • 302bd51d17 Improve online chat actor prompt for research and normal mode Debanjum 2024-10-31 17:56:06 -0700
  • 52163fe299 Improve research planner prompt to reduce looping Debanjum 2024-10-31 18:15:58 -0700
  • 7ebf999688 Merge branch 'master' of github.com:khoj-ai/khoj into features/advanced-reasoning sabaimran 2024-10-31 18:15:13 -0700
  • 159ea44883 Remove frame references in the diagramming prompts sabaimran 2024-10-31 18:14:51 -0700
  • 89597aefe9 Json dump contents in prompt tracer to make structure discernable Debanjum 2024-10-31 18:04:43 -0700
  • 5b15176e20 Only add /research prefix in research mode if not already in user query Debanjum 2024-10-31 18:02:55 -0700
  • 559601dd0a Do not exit if/else loop in research loop when notes not found sabaimran 2024-10-31 13:51:10 -0700
  • a13760640c Only show trash can when turnId is present sabaimran 2024-10-31 13:19:16 -0700
  • 8d1ecb9bd8 Add optional brew steps for docker install sabaimran 2024-10-31 12:41:53 -0700
  • adca6cbe9d Merge branch 'master' into add-prompt-tracer-for-observability Debanjum 2024-10-31 02:28:34 -0700
  • e17dc9f7b5 Put train of thought ui before Khoj response on web app Debanjum 2024-10-31 02:24:53 -0700
  • e8e6ead39f Fix deleting new messages generated after conversation load Debanjum 2024-10-30 18:10:43 -0700
  • cb90abc660 Resolve train of thought component needs unique key id error on web app Debanjum 2024-10-30 05:03:36 -0700
  • ca5a6831b6 Add ability to delete messages from the web app Debanjum 2024-10-30 04:24:22 -0700
  • ba15686682 Store turn id with each chat message. Expose API to delete chat turn Debanjum 2024-10-30 04:21:58 -0700
  • f64f5b3b6e Handle add/delete file filter operation on non-existent conversation Debanjum 2024-10-30 02:23:34 -0700
  • b3a63017b5 Support setting seed for reproducible LLM response generation Debanjum 2024-10-30 02:20:38 -0700
  • d44e68ba01 Improve handling embedding model config from admin interface Debanjum 2024-10-29 13:15:47 -0700
  • 358a6ce95d Defer turning cursor color to selected agents color for later Debanjum 2024-10-28 20:49:01 -0700
  • 2ac840e3f2 Make cursor in chat input take on selected agent color Debanjum 2024-10-28 16:26:23 -0700
  • 1448b8b3fc Use 3rd person for user in research prompt to reduce person confusion Debanjum 2024-10-29 20:56:36 -0700
  • b8c6989677 Separate example from actual question in extract question prompt Debanjum 2024-10-29 11:40:03 -0700
  • 86ffd7a7a2 Handle \n, dedupe json cleaning into single function for reusability Debanjum 2024-10-29 11:47:11 -0700
  • 83ca820abe Encourage Anthropic models to output json object using { prefill Debanjum 2024-10-29 11:22:29 -0700
  • dc8e89b5de Pass tool AIs iteration history as chat history for better context Debanjum 2024-10-28 19:21:20 -0700
  • d865994062 Rename code tool arg previous_iteration_history' to context' Debanjum 2024-10-28 21:12:16 -0700
  • 06aeca2670 Make researcher, docs search AIs ask more diverse retrieval questions Debanjum 2024-10-28 20:43:53 -0700
  • 01881dc7a2 Revert "Make extract question prompt in 1st person wrt user as its a user message" Debanjum 2024-10-29 11:35:07 -0700
  • 3e695df198 Make extract question prompt in 1st person wrt user as its a user message Debanjum 2024-10-28 20:41:36 -0700
  • a3751d6a04 Make extract relevant information system prompt work for any document Debanjum 2024-10-28 20:39:10 -0700
  • a39e747d07 Improve passing user name in pick next research tool prompt Debanjum 2024-10-28 20:29:47 -0700
  • deff512baa Improve research mode prompts to reduce looping, increase webpage reads Debanjum 2024-10-28 15:37:54 -0700
  • d3184ae39a Simplify storing and displaying document results in research mode Debanjum 2024-10-28 20:26:59 -0700
  • 8bd94bf855 Do not use a message branch if no msg id provided to prompt tracer Debanjum 2024-10-28 17:48:45 -0700
  • b63fbc5345 Add a simple badget to the dropdown menu that shows subscription status sabaimran 2024-10-30 13:00:16 -0700
  • 82f3d79064 Merge branch 'master' of github.com:khoj-ai/khoj into features/advanced-reasoning sabaimran 2024-10-30 11:32:10 -0700
  • 2b2564257e Handle subscription case where it's set to trial, but renewal_date is not set. set the renewal_date for LENGTH_OF_FREE_TRIAL days from subscription creation. sabaimran 2024-10-30 11:05:31 -0700
  • 9935d4db0b Do not use a message branch if no msg id provided to prompt tracer Debanjum 2024-10-28 17:48:45 -0700
  • d184498038 Pass context in separate message from user query to research chat actor Debanjum 2024-10-28 05:59:32 -0700
  • d75ce4a9e3 Format online, notes, code context with YAML to be legibile for LLM Debanjum 2024-10-28 05:56:31 -0700
  • 5bea0c705b Use break-words in the train of thought for better formatting sabaimran 2024-10-28 15:36:06 -0700
  • 1f1b182461 Automatically carry over research mode from home page to chat sabaimran 2024-10-28 15:29:24 -0700
  • ebaed53069 Merge branch 'master' of github.com:khoj-ai/khoj into features/advanced-reasoning sabaimran 2024-10-28 12:39:00 -0700
  • 889dbd738a Add keyword diagram to diagram output mode description sabaimran 2024-10-28 12:20:46 -0700
  • 50ffd7f199 Merge branch 'master' into features/advanced-reasoning Debanjum 2024-10-28 04:07:36 -0700
  • a5d0ca6e1c Use selected agent color to theme the chat input area on home page Debanjum 2024-10-28 01:19:37 -0700
  • aad7528d1b Render slash commands popup below chat input text area on home page Debanjum 2024-10-28 01:55:24 -0700
  • 3e17ab438a
    Separate notes, online context from user message sent to chat models (#950) Debanjum 2024-10-28 02:03:18 -0700
  • 8ddd70f3a9 Put context into separate message before sending to offline chat model Debanjum 2024-10-28 00:22:21 -0700
  • ee0789eb3d Mark context messages with user role as context role isn't being used Debanjum 2024-10-27 23:55:47 -0700
  • 4e39088f5b Make agent name in home page carousel not text wrap on mobile Debanjum 2024-10-27 23:02:27 -0700
  • 94074b7007 Focus chat input on toggle research mode. v-align it with send button Debanjum 2024-10-27 22:54:55 -0700
  • a691ce4aa6 Batch entries into smaller groups to process sabaimran 2024-10-27 20:43:41 -0700
  • 2924909692 Add a research mode toggle to the chat input area sabaimran 2024-10-27 16:37:40 -0700
  • 68499e253b Auto-collapse train of thought, show after chat response in history sabaimran 2024-10-27 15:48:13 -0700
  • 101ea6efb1 Add research mode as a slash command, remove from default path sabaimran 2024-10-27 15:47:44 -0700
  • 0bd78791ca Let user exit from command mode with esc, click out, etc. sabaimran 2024-10-27 15:01:49 -0700
  • a121d67b10 Persist the train of thought in the conversation history sabaimran 2024-10-26 23:46:15 -0700
  • 9e8ac7f89e Fix input/output mismatches in the /summarize command sabaimran 2024-10-26 16:37:58 -0700
  • e4285941d1 Use the advanced chat model if the user is subscribed sabaimran 2024-10-26 16:00:54 -0700
  • 33e48aa27e Merge branch 'add-prompt-tracer-for-observability' of github.com:khoj-ai/khoj into features/advanced-reasoning sabaimran 2024-10-26 14:09:00 -0700
  • fd71a4b086 Add better exception handling in the prompt trace logic, use default value from parameters sabaimran 2024-10-26 14:08:00 -0700
  • 3e5b5ec122 Encourage model to read webpages more often after online search Debanjum 2024-10-26 10:41:42 -0700
  • bf96d81943 Format online results as YAML to pass it in more readable form to model Debanjum 2024-10-26 10:39:45 -0700
  • 3e97ebf0c7 Unescape special characters in prompt traces for better readability Debanjum 2024-10-26 10:45:42 -0700
  • 8af9dc3ee1 Unescape special characters in prompt traces for better readability Debanjum 2024-10-26 10:45:42 -0700
  • 0f3927e810 Send gathered references to client after code results calculated Debanjum Singh Solanky 2024-10-26 05:59:10 -0700
  • f04f871a72 Merge branch 'add-prompt-tracer-for-observability' of github.com:khoj-ai/khoj into features/advanced-reasoning Debanjum Singh Solanky 2024-10-26 05:30:24 -0700
  • ddc6ccde2d Merge branch 'master' into features/advanced-reasoning Debanjum Singh Solanky 2024-10-26 05:15:51 -0700
  • ea0712424b Commit conversation traces using user, chat, message branch hierarchy Debanjum Singh Solanky 2024-10-23 20:02:28 -0700
  • a3022b7556 Allow Offline Chat model calling functions to save conversation traces Debanjum Singh Solanky 2024-10-24 14:26:57 -0700
  • eb6424f14d Allow Anthropic API calling functions to save conversation traces Debanjum Singh Solanky 2024-10-24 14:26:30 -0700
  • 6fcd6a5659 Allow Gemini API calling functions to save conversation traces Debanjum Singh Solanky 2024-10-24 14:15:53 -0700