Skip to content

Releases: fuchicar/auris-ai

v1.3.1

Choose a tag to compare

@github-actions github-actions released this 30 Jul 17:19
94722c4

Changelog

Bug fixes

  • cf9d572 Revert "fix(market): redact API key from transport errors before they reach the LLM"
  • 5b0733f fix(agent): cascade market chain on successful-but-empty results (#30)
  • a34ff1f fix(agent): make tax P&L period bounds end-of-day inclusive and UTC-consistent (#33)
  • b543345 fix(agent,drivers): surface abnormal stream termination instead of a silent fake stop (#39)
  • 2a96a76 fix(config,portfolio): add entropy to timestamp IDs to prevent same-second collisions
  • b4576ec fix(market): redact API key from transport errors before they reach the LLM
  • 8b1740b fix(market): redact API key from transport errors before they reach the LLM
  • 640a36d fix(tui): adaptive panel width + scrollable disclaimer viewport (issue #37)
  • 3e9bb01 fix(tui): confirm before -setup overwrites an existing config, clarify unlock errors
  • 438a0e5 fix(tui): gofmt screen_disclaimer.go and its test (issue #37)
  • 7cfebe0 fix(tui): manage vertical height in instrument and portfolio views (issue #36)
  • 53e2aa5 fix(tui): preserve portfolio's AI provider/model when editing
  • a7dc193 fix(tui): re-inject terminal size into screens created after startup
  • 78d74cc fix(tui): restore provider config on cancelled /aiproviders or /marketproviders (#38)
  • 498633c fix(tui): review fixes for #37 (theme preview width, disclaimer key dispatch, dead code, gofmt)
  • a4795b2 fix(tui): windowed rendering + accurate chrome budgets for long lists (issue #35)

Other changes

  • a2bbc37 test(tui): sandbox every pkg/tui test behind an isolated XDG_CONFIG_HOME

v1.3.0

Choose a tag to compare

@github-actions github-actions released this 28 Jul 16:35
f405434

Changelog

Features

  • 419bc95 feat: add promptlab, a system-prompt testing/refinement harness

Bug fixes

  • 27504cf fix(agent): portfolio_add_lot no longer debits cash by default
  • 8e4efd1 fix(portfolio): catalogued lots no longer inflate benchmark returns
  • 4ca3b85 fix(tui): don't discard AgentModel via Shift+Tab while streaming
  • 716ad7f fix(tui): let Esc exit the API-key screen from /marketproviders
  • 10b55b8 fix(tui): reload portfolio from disk before /new, session select, and resolvePortfolioSession
  • 4986c1c fix(tui): reset flowContext on every exit from portfolio-agent mode
  • 0467631 fix: reject wrong passphrase even when no API keys are stored (#23)

Other changes

  • 4d09397 ci: run build/vet/test on PRs and pushes to develop
  • 19317d5 test(news): don't fail CI on transient upstream errors in live feed check

v1.2.0

Choose a tag to compare

@github-actions github-actions released this 21 Jul 16:33
242f9ac

Changelog

Features

  • 871eea3 feat: harden agent against prompt-manipulation attempts (#8)

Bug fixes

  • 1e62892 fix: reload portfolio from disk when /menu returns from portfolio-agent mode

Other changes

  • 32de5de Refactor code structure for improved readability and maintainability
  • 4a4bde5 chore: move cmd/styletest to tools/styletest

v1.1.0

Choose a tag to compare

@github-actions github-actions released this 17 Jul 09:27
68b2709

Changelog

Features

  • 6bdb097 feat: add contextual explanations to setup wizard screens

Bug fixes

  • e578bd3 fix: support multiple named OpenAI-Compatible AI providers
  • e5d8960 fix: wrap error messages instead of overflowing the terminal

v1.0.0

Choose a tag to compare

@github-actions github-actions released this 15 Jul 10:16

Changelog

Features

  • 74b04a9 feat(math): enhance error handling and documentation for financial calculations
  • f67a946 feat(math): implement financial calculation tools for P/FCF, PEG, dividend yield, and growth; update tests and tool registration
  • 7599442 feat(math): implement stress test calculation tool; add tests and update tool registration
  • 2725159 feat(minimax): add MiniMax driver implementation and tests
  • 065bcf5 feat(minimax): add documentation for ListModels method in MiniMax driver
  • be372b2 feat(monte_carlo): implement Monte Carlo simulation tool and associated tests
  • 129831e feat(news): update ToolDescription and parameter descriptions to English
  • 686bc8f feat(ollama): improve context handling and timeout management in Stream and Complete methods
  • 7e77f0e feat(portfolio): add buy action and success message for portfolio instruments
  • 0c36441 feat(portfolio): add cash management features and update portfolio metrics
  • 08f3eec feat(portfolio): add portfolio management tools and integrate with agent model
  • c45c97b feat(portfolio): enhance transaction handling and add new tools
  • 6d9d36f feat(portfolio): implement portfolio_suggest_rebalance tool and associated logic for suggesting buy/sell operations
  • 3a9955a feat(portfolio): implement target allocation management; add UI and backend support
  • 7fc7b7d feat(prompts): add comprehensive guide for writing system prompts and update existing prompt structure to English
  • 758a912 feat(transactions): implement transaction history screen and related functionality
  • b3d65fc feat(tui): enhance instrument search model with dynamic sizing and scrolling support
  • 779729f feat(tui): implement portfolio instruments and menu screens
  • 327e385 feat(tui): implement user interface for setup and configuration
  • fa938eb feat: Implement comprehensive numeric input validation for financial calculations
  • b6fb7a1 feat: Update ADRs and complete Tier C tasks
  • f5d3bdf feat: add AI usage disclaimer and warning messages for financial decisions
  • 5d6896f feat: add Anthropic Claude driver and integration tests
  • 07cd3fd feat: add ComputedAt field to financial calculation results
  • 701a60f feat: add English and Spanish README files with installation and usage instructions
  • 658a694 feat: add MIT License to the project
  • f6396ac feat: add Nix configuration files for development environment
  • 8192f93 feat: add Ollama driver for local LLM inference
  • 4f16d00 feat: add agent mode and configuration screens
  • 0eb97d4 feat: add candlestick chart and sparkline support in TUI, implement SMA calculation, and enhance portfolio views
  • 269e504 feat: add configuration menu options and enhance navigation in the main menu
  • 57a27c4 feat: add context window reporting and info command for agent
  • 46ca151 feat: add currency formatting and selection in portfolio management
  • 27cdd8f feat: add edit profile option to menu and enhance profile management
  • 42b7b3d feat: add integration tests for Ollama driver and model configuration
  • 6d2179e feat: add new theme options and enhance agent message rendering styles
  • 429ddb3 feat: add news feeds management functionality and UI
  • 6a7b773 feat: add portfolio benchmark comparison tool and related functionality
  • 7c0817a feat: add progress tracking for tool execution and enhance model selection interface
  • b35e060 feat: add simulation mode for market data
  • 12d4bf2 feat: add time tools for current, today's, and yesterday's date retrieval
  • cb6abae feat: add unit tests for math tools and enhance system message prompts
  • df6d023 feat: add watchlist functionality with live price updates and % change display
  • 84f960d feat: enhance ProgressEvent handling to surface distinct tool calls and improve deduplication logic
  • 79dfc56 feat: enhance adaptive color handling for agent UI elements and improve label rendering
  • c46fc81 feat: enhance agent connection by integrating market provider API
  • b04b0b9 feat: enhance color handling and background persistence in agent rendering
  • 5be2655 feat: enhance documentation and metrics, add atomic file write utility
  • cce15bd feat: enhance financial metrics with DividendYieldTTM and Beta, update handling for missing parameters in calculations
  • 0b36be5 feat: enhance market provider management and localization
  • efd4300 feat: enhance system prompts with mathematical expression guidelines and rendering rules
  • 1b40728 feat: implement Agent structure with chat functionality and tool dispatching
  • fad5795 feat: implement at-rest encryption for portfolios and chat sessions
  • f9429eb feat: implement back navigation in setup wizard and enhance UI indicators
  • 433c9cf feat: implement capability filtering for agent tools based on provider support
  • cffdd40 feat: implement change passphrase functionality with validation and UI updates
  • 120296d feat: implement market_render_price_chart tool and integrate chart rendering in TUI
  • ac19044 feat: implement mutex protection for lastUsage and storageKey to prevent data races
  • 42e8e11 feat: implement portfolio export functionality to JSON and CSV formats
  • e7536af feat: implement quotes snapshot handling in portfolio metrics and enhance prompts for technical analysis
  • 6018928 feat: implement session management with session picker and migration from ChatHistory
  • abc338f feat: implement streaming support for agent interactions
  • f7afa98 feat: mask API key inputs with EchoPassword for enhanced security
  • e404647 feat: replace text input with textarea for improved user experience and dynamic height adjustment
  • 137c0ff feat: surface error messages when model loading fails and reset pending states
  • e78413d feat: update dependencies and enhance Markdown rendering in agent prompts
  • ee64a41 feat: update renderer style in ensureRenderer function to use standard theme
  • 87e782a feat: update toolset to include Sortino, Treynor, and Information Ratio calculations

Bug fixes

  • f1b9350 fix(docs): correct usage flag notation in help text
  • 41e73be fix(locale): add applyStoredLocale method to initialize translations from config
  • 6d973fc fix: enhance exit condition in agent loops to handle empty tool calls
  • 40e94c0 fix: improve error handling for numeric arguments to prevent silent coercion to zero
  • a80a636 fix: preserve assistant text alongside tool calls in OpenAI driver responses
  • 9a852d1 fix: replace hardcoded placeholders with localized strings for improved internationalization
  • 20b05f0 fix: update portfolio_add_instrument to include optional debit_cash parameter for cash transactions

Other changes

  • 08a2ef8 "Claude Code Review workflow"
  • d599958 "Claude PR Assistant workflow"
  • b3a5f2e Add FMP driver tests and provider API definitions
  • a95a025 Add Ollama and Gemini AI providers with integration tests
  • 7144634 Add OpenAI driver and integration tests
  • 9baebd6 Add financial calculation tools and corresponding tests
  • cb782e6 Add financial validation functions and calculations
  • c97c2dd Add integration tests for agent functionality and improve agent prompts
  • cfd9aff Add portfolio tax P&L calculation and related tests
  • 7a695d4 Add presentation HTML for TFM project with styled slides and interactive features
  • c684ffc...
Read more