Ling0925
/
buster
mirror of https://github.com/buster-so/buster.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity
655 Commits 144 Branches 17 Tags 111 MiB
Commit Graph

205 Commits

Author SHA1 Message Date
dal 8fedcecd43
handlers lib 2025-02-14 13:44:21 -07:00
dal a59e9a26c2
Remove redundant foreign key constraints and joinables for messages_to_files 2025-02-14 13:32:50 -07:00
dal 743c256dbc
added in some testing work and changed some rules 2025-02-14 11:32:33 -07:00
dal 51b6d00159
message fkey error still propagating. 2025-02-14 08:32:06 -07:00
dal da8eb26931
silent on parsing errors since we will have a ton with the optimistic parsing. 2025-02-13 16:36:04 -07:00
dal eac2529f29
ok we are on our way there... 2025-02-13 16:28:59 -07:00
dal 3bd5516486
ok thread is being inserted 2025-02-13 16:15:09 -07:00
dal 977b5eb6de
mdc 2025-02-13 15:05:10 -07:00
dal 2d47c2c539
mdc commit 2025-02-13 14:50:24 -07:00
dal bdeaccdca8
cursor rules 2025-02-13 13:31:53 -07:00
dal 9fc5011b75
cursor rules 2025-02-13 08:38:14 -07:00
dal 5416418955
cargo optimization for faster builds 2025-02-13 08:31:14 -07:00
dal 041859ec28
change thought_pills to thoughts 2025-02-11 15:45:05 -07:00
dal bf24b249d4
Refactor message and thread handling with explicit UUID references 2025-02-11 12:46:21 -07:00
dal 8e8140246d
Update message transformation to include chat and message IDs 2025-02-11 12:34:06 -07:00
dal 8b96ec01fb
Refactor message transformation with new container types and UUIDs 2025-02-11 11:53:06 -07:00
dal bf05c7f06b
Add new thread events for chat generation progress tracking 2025-02-11 11:21:57 -07:00
dal e7b96d9bd5
make a chart card 2025-02-11 10:58:25 -07:00
dal 41a7146a8d
Add initial chat response for new thread creation 2025-02-11 10:05:20 -07:00
dal 6fcfc4fba6
ids matching for stream messages 2025-02-11 09:57:19 -07:00
dal 2376153459
ids and initial message repeat handled 2025-02-11 09:36:28 -07:00
dal 973e9b41ce
ids are set 2025-02-11 09:21:08 -07:00
dal fcc15a0e6f
fix tool name transform path 2025-02-11 08:23:58 -07:00
dal 8940fbf3b6
using name as id for file type reasoning message 2025-02-11 08:21:32 -07:00
dal 8de08323fa
consistent message id for text stream 2025-02-11 08:10:58 -07:00
dal bcf1ac1a65
file stream is working 👍🏼 for create file. 2025-02-11 07:49:22 -07:00
dal afb40c90be
ok restructured the transformer to get vecs back 2025-02-11 07:40:07 -07:00
dal 90931fc029
Add support for chat routes and update thread route handling 2025-02-11 07:27:56 -07:00
dal d75931dcb0
We did it joe 2025-02-11 07:14:06 -07:00
dal 43e2cf44f4
its working, but not sure if it's how I want it... 2025-02-10 17:09:01 -07:00
dal 61153020ba
save point 2025-02-10 16:08:08 -07:00
dal 8c6c1d83ee
still working on the darn stream 2025-02-10 14:49:22 -07:00
dal fd65408e30
clean up to try different impl 2025-02-10 14:22:13 -07:00
dal 2593536efa
So close, just a few more tweaks to clean up stream. 2025-02-10 13:35:57 -07:00
dal 118ef9c691
ok streaming it back 2025-02-10 12:29:13 -07:00
dal 8c8372b50e
ok everything sending back except create and modify 2025-02-10 12:15:21 -07:00
dal d0400b5226
transforms for events 2025-02-10 11:53:19 -07:00
dal 233b580e1c
data catalog search transformation 2025-02-10 11:29:08 -07:00
dal 6a67931667
message transform 2025-02-10 10:47:16 -07:00
dal 93206fcf5c
added in duration and passed back req params 2025-02-10 09:21:07 -07:00
dal 9169a1c9e8
small tweaks 2025-02-10 08:26:25 -07:00
dal c849a22b4f
added in progress on tools 2025-02-10 08:24:56 -07:00
dal 8bd14e0ee7
added in progress on stream messages 2025-02-10 08:17:14 -07:00
dal 789b22fe1e
Improve message streaming and tool call processing 
- Enhance message streaming with more precise content and tool call handling
- Add logic to only send and store meaningful assistant messages and tool calls
- Prevent sending empty or redundant messages during stream processing
- Improve tool call and content update tracking in agent stream method
- Optimize message inclusion in recursive thread generation
 2025-02-10 07:30:52 -07:00
dal 641527114c
refactor(agent): Enhance LLM request handling with improved tool call processing 
- Add support for initial non-tool response and subsequent tool-enabled processing
- Implement `PendingToolCall` to manage incremental tool call updates
- Update `Delta` and related types to support more flexible tool call streaming
- Modify agent stream processing to handle tool calls with improved state management
- Add robust handling for tool call deltas and function call arguments
 2025-02-10 07:17:14 -07:00
dal 373a83efea
refactor(agent): Extract agent prompt to constant and update date 
- Move hardcoded agent prompt to a constant `AGENT_PROMPT` in the `agent_thread.rs` file
- Update the prompt's date from January 27, 2025 to February 7, 2025
- Simplify thread initialization by referencing the new constant
- Maintain existing prompt structure and guidelines
 2025-02-07 14:18:02 -07:00
dal 4b743fe5ec
feat(agent): Add recursion depth limit to prevent infinite processing 
- Implement a maximum recursion depth of 30 for agent thread processing
- Add recursion depth tracking to prevent potential infinite loops
- Provide user-friendly message when maximum recursion depth is reached
- Update debug logging to include current recursion depth
- Modify both synchronous and streaming thread processing methods
 2025-02-07 11:35:13 -07:00
dal 8b51618afd
refactor(agent): Implement recursive stream processing with improved tool execution 
- Refactor agent stream processing to use a recursive approach for handling tool calls
- Enhance tool execution with more robust error handling and result tracking
- Improve stream chunk processing with detailed state management
- Add support for recursive thread generation based on tool call results
- Implement cloning for LiteLLMClient to support stream processing tasks
 2025-02-07 11:22:43 -07:00
dal bb4e4ca9d8
refactor(ai_tools): Standardize search tool response format and requirements 
- Update search data catalog and file search tools to use consistent JSON response structure
- Enforce strict response format with mandatory "results" key and comprehensive field requirements
- Add explicit guidelines for result composition and handling of empty result sets
- Improve response predictability and parsing reliability for AI-powered search tools
 2025-02-07 11:12:08 -07:00
dal 8054bedf1a
fix(ai_tools): Update LLM request parameters and improve response handling 
- Add `stream: Some(false)` to file search and data catalog tools
- Make `json_schema` optional in `ResponseFormat` serialization
- Enhance logging in search file tool with debug and warning messages
- Improve error context when parsing LLM JSON responses
 2025-02-07 11:05:39 -07:00