shokollm
caef4b36ed
feat: auto-fill token from strategy config in backtest page
2026-04-11 04:37:52 +00:00
shokollm
3bf2877df2
fix: append -bsc suffix to token address for klines API
2026-04-10 17:07:14 +00:00
shokollm
145c6710d1
fix: set 1-day range for backtest (start day before end day)
2026-04-10 16:30:22 +00:00
shokollm
3c8c85aefc
fix: table regex to match tables anywhere in text (not just at start)
2026-04-10 13:53:25 +00:00
shokollm
39b2b558a5
fix: export parseInlineElements and types from markdown.ts
2026-04-10 13:17:07 +00:00
shokollm
7795753aaa
fix: render bold and inline code formatting in list items
2026-04-10 13:14:17 +00:00
shokollm
36dcfdb6e2
chore: restrict agent to BSC only, remove chain parameter from search_tokens tool
2026-04-10 12:58:25 +00:00
shokollm
48fc323dac
fix: handle native tool_calls from MiniMax API instead of parsing JSON from content
2026-04-10 12:54:29 +00:00
shokollm
0af2de7209
feat: add search_tokens tool for AI to recommend trending tokens
2026-04-10 12:48:49 +00:00
shokollm
e82b8b3549
fix: update token search to use trending endpoint (v2/tokens doesn't exist)
2026-04-10 12:37:27 +00:00
shokollm
6f23b322d3
feat: add token search in modal when confirming address
2026-04-10 12:14:32 +00:00
shokollm
297a185215
feat: implement token address confirmation dialog and limit backtest duration
2026-04-10 11:52:40 +00:00
shokollm
f86ff75525
fix: remove extra closing brace in CSS
2026-04-10 11:32:11 +00:00
shokollm
6f9564790f
docs: add ISSUES.md for tracking open issues
2026-04-10 11:13:20 +00:00
shokollm
f43eb11f6f
feat: improve backtest with manual refresh and token address confirmation
2026-04-10 10:54:42 +00:00
shokollm
446da96ce4
fix: search for token first to get proper token_id before fetching klines
2026-04-10 10:47:33 +00:00
shokollm
922ef89c1e
feat: add backtest progress tracking and fix stop functionality
2026-04-10 10:43:04 +00:00
shokollm
a601ebb08b
fix: handle datetime serialization in backtest and show errors in frontend
2026-04-10 10:34:29 +00:00
shokollm
bb40193fc3
fix: add required chain field (bsc) to backtest request
2026-04-10 10:28:16 +00:00
shokollm
3a7d3a3732
feat: set default dates for backtest (yesterday to 30 days ago)
2026-04-10 10:23:35 +00:00
shokollm
0f558a5e8e
fix: handle array error format from FastAPI validation errors
2026-04-10 10:21:27 +00:00
shokollm
9e9ff6fa7f
fix: handle undefined timeframe in strategy preview
2026-04-10 10:19:28 +00:00
shokollm
4c48932ece
fix: support inline formatting in table cells (bold, italic, code, links)
2026-04-10 10:15:21 +00:00
shokollm
bfc85648db
fix: improve markdown parser for tables, headings, and line breaks
2026-04-10 10:09:46 +00:00
shokollm
925920eee1
fix: add typing indicator back when waiting for response
2026-04-10 10:05:50 +00:00
shokollm
299e74cffa
chore: hide ProUpgradeBanner (not implementing pro yet)
2026-04-10 09:59:08 +00:00
shokollm
2b875cfa27
feat: show thinking above response with expand/collapse, first line preview
2026-04-10 09:56:21 +00:00
shokollm
ae612ad725
fix: use requests instead of OpenAI client for thinking endpoint
2026-04-10 09:50:36 +00:00
shokollm
08912019c2
feat: use MiniMax extended thinking endpoint for native reasoning
2026-04-10 09:47:09 +00:00
shokollm
44453877b3
feat: use direct LLM with structured JSON for thinking/response separation
2026-04-10 09:31:07 +00:00
shokollm
ad4a1e89d5
fix: revert to kickoff (stream not available on Agent)
2026-04-10 09:23:46 +00:00
shokollm
57fa200ba9
feat: add thinking content to chat response
2026-04-10 09:16:08 +00:00
shokollm
db4fb83243
feat: add markdown rendering and thinking state UI to chat
2026-04-10 09:01:16 +00:00
shokollm
560b61c431
fix: increase timeout for long-running AI chat requests
2026-04-10 08:51:03 +00:00
shokollm
c6baadf8b8
fix: use JSON body for login instead of form data
2026-04-10 08:09:42 +00:00
shokollm
937cc2da60
fix: send username instead of email for login API
2026-04-10 07:47:21 +00:00
32cd7184ea
Merge pull request 'feat: implement conversational AI agent with tool-calling' ( #52 ) from feat/conversational-agent-with-tools into main
2026-04-10 09:39:45 +02:00
shokollm
765e390b9b
feat: implement conversational AI agent with tool-calling
...
Prototype implementation that allows:
1. Normal conversation with the AI
2. Tool-calling to update trading strategies
Created new ConversationalAgent that uses CrewAI with tools:
- get_current_strategy: Check current bot strategy
- update_trading_strategy: Update bot's trading configuration
The agent can now respond to questions like 'What is this?' without
forcing JSON output, and can update strategies when user provides
specific parameters.
Refs #51
2026-04-10 05:00:22 +00:00
21ce282cae
Merge pull request 'fix: add fallback UUID generator for crypto.randomUUID compatibility' ( #50 ) from fix/crypto-randomuuid-fallback into main
2026-04-10 06:26:49 +02:00
shokollm
4fa9b0456a
fix: add fallback UUID generator for crypto.randomUUID compatibility
...
crypto.randomUUID() is not available in all environments (e.g., older browsers,
non-secure contexts). Added a fallback UUID v4 implementation.
2026-04-10 04:19:45 +00:00
af9900d0ba
Merge pull request 'fix: add timeout for chat requests and improve error handling' ( #49 ) from fix/chat-timeout-handling into main
2026-04-10 06:15:45 +02:00
shokollm
b3ab004447
fix: add timeout for chat requests and improve error handling
...
Changes:
1. Add 30-second timeout for chat API requests using AbortController
2. User's message now shows immediately before API response (already done in previous PR)
3. Differentiate between timeout errors and other errors in error messages
4. API client now accepts optional signal parameter for abort support
2026-04-10 04:09:30 +00:00
d394bc0857
Merge pull request 'fix: display user messages in chat interface' ( #48 ) from fix/display-user-messages into main
2026-04-10 06:04:23 +02:00
shokollm
dfa806ab53
fix: add user's message to frontend chat store when sending
...
Previously, only the assistant's response was added to the frontend store.
Now both user and assistant messages are stored, so the conversation
displays correctly in the chat interface.
2026-04-10 04:00:51 +00:00
3493775b7f
Merge pull request 'fix: update MiniMaxConnector default model to MiniMax-M2.7' ( #47 ) from fix/minimax-connector-model into main
2026-04-10 06:00:36 +02:00
shokollm
82645dfb3b
fix: update MiniMaxConnector default model to MiniMax-M2.7
2026-04-10 03:53:26 +00:00
c17fa243a1
Merge pull request 'fix: use MiniMax text/chatcompletion_v2 endpoint' ( #46 ) from fix/minimax-endpoint-v2 into main
2026-04-10 05:44:25 +02:00
shokollm
a55ed9cc04
fix: use MiniMax text/chatcompletion_v2 endpoint instead of chat/completions
...
The /v1/chat/completions endpoint returns 529 (overloaded) while
/v1/text/chatcompletion_v2 works reliably.
2026-04-10 03:42:20 +00:00
d1408b74b4
Merge pull request 'fix: properly configure MiniMax API endpoint and CrewAI LLM' ( #45 ) from fix/minimax-api-endpoint-v2 into main
2026-04-10 05:31:13 +02:00
shokollm
4197475eed
fix: properly configure CrewAI LLM with MiniMax api_base
...
- Use CrewAI's LLM class directly with api_base parameter instead of custom subclass
- Remove broken MiniMaxLLM inheritance from LLM
- Update agent creation to use LLM(model, api_key, api_base) pattern
The issue was that inheriting from CrewAI's LLM class caused the api_base
to be set to None. Now we use CrewAI's LLM directly with the correct parameters.
Fixes #43
2026-04-10 03:19:51 +00:00
