Split conversational.py (2271 lines) into modular files:
- tools.py: TOOL_REGISTRY, get_tool_registry(), SKILL_EMOJIS
- help.py: format_* functions for slash command help
- client.py: MiniMaxClient, SYSTEM_PROMPT, TOOLS definitions
- agent.py: ConversationalAgent class with all methods
- __init__.py: Public exports from all modules
Updated bots.py import to use new module path.
Deleted conversational.py.
- Display token name and symbol in risk analysis output
- Handle is_honeypot: -1 as 'Unknown (could not determine)'
- Show risk level (Low/Medium/High) with risk score
- Use risk_level field instead of status
- Store recent search results in agent instance
- When price returns empty, suggest using /search tool
- Check if user input matches recent search results and use that address
The price API requires full contract addresses (0x...-bsc), but users typically provide symbols.
Now /price TRADOOR will search for the token and show price info from search results.
- Added _is_error_output helper to detect errors in CLI output
- API errors like 'API error 403' now show proper error message instead of 'No price data available'
- Updated all command execution methods to use the helper
- Added _execute_backtest_direct() that extracts params from message or strategy config
- Added _execute_simulate_direct() that extracts params from message or strategy config
- Both commands now execute directly when called with /backtest or /simulate
- If token address is missing, asks user for the parameter
Commands now execute in two steps:
1. User sends /search -> acknowledge and wait for param
2. User sends 'pepe' -> auto-execute search with 'pepe'
Commands without params (/backtest, /simulate, /trending, /strategy) execute directly.
Pending commands tracked via self.pending_command state.
- Added _execute_trending method that runs the trending CLI command
- Returns formatted list of trending tokens on BSC
- Shows error message if no tokens found or command fails
- Added _get_strategy_response method to query bot's strategy_config from DB
- Shows formatted strategy with conditions, actions, and risk management
- Shows helpful message if no strategy is configured yet
- Reset conversational.py to pr-58 working AVE integration
- Added TOOL_REGISTRY with all available slash commands
- Added _handle_slash_command for skill activation
- Slash commands show brief acknowledgment when used alone
- Slash commands with args are passed to AI for handling
- Added dropdown menu in ChatInterface for skill discovery
- Menu positions above textarea
- Menu shows filtered tools as user types
Implement slash command help system as described in issue #57:
- Add Tool Registry in backend with metadata for all available tools
- Add command parser for '/' prefix in ConversationalAgent
- Add slash command handling functions:
- '/' shows list of all available tools
- '/help' shows general help about Randebu
- '/<tool-name>' shows detailed help for specific tool
- Update frontend ChatInterface to detect '/' and show formatted help dropdown
- Add keyboard navigation (Arrow keys, Tab, Enter, Escape) for slash menu
Instead of importing library functions directly, now calling the
ave_data_rest.py CLI script via subprocess. This follows the
recommended approach from the SKILL.md documentation.
Changes:
- Add _call_ave_script helper method for subprocess calls
- Update search_tokens, get_token, get_price, get_risk to use CLI
- Set AVE_USE_DOCKER=false to run scripts directly without Docker
- Remove direct imports of ave.http module
Agent can now manage simulations via conversation:
- 'start' (run simulation on token)
- 'stop' (stop running simulation)
- 'status' (check if running, show progress)
- 'results' (get current/latest simulation results)
Example conversations:
- 'Run a simulation on GRIAN' → starts simulation
- 'How's the simulation going?' → shows status
- 'Stop the simulation' → stops and shows results
- 'Show me the results' → shows simulation results
Agent can now run backtests via conversation:
- User can say 'backtest this strategy' or 'test on [token]'
- Agent extracts token address (from conversation or asks user)
- Agent runs backtest and presents results in chat with:
- Total Return
- Final Balance
- Win Rate
- Max Drawdown
- Sharpe Ratio
- Agent suggests strategy adjustments based on results
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
- 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
Changes:
1. Updated API endpoint from api.minimax.chat to api.minimax.io
2. Changed default model from MiniMax-Text-01 to MiniMax-M2.7
(MiniMax-Text-01 is not available for all API key plans)
3. Updated .env.example with correct default model
MiniMax API docs: https://platform.minimax.io/docs/api-reference/text-anthropic-apiFixes#43
- Fix relative import path in crew.py (from ..core to ...core)
- Update __init__.py exports to match actual class names
- Remove incorrect CrewAgent and LLMConnector exports
LLM was outputting nested params structure but engines expect flat fields.
This caused backtesting and simulation to never trigger any trades.
Changes:
- llm_connector.py: Update prompt to output flat condition structure
- crew.py: Update StrategyValidator to validate flat structure
- crew.py: Update StrategyExplainer to read flat structure
Fixes#25
