aipa-cli 0.1.4__tar.gz → 0.1.6__tar.gz

{aipa_cli-0.1.4 → aipa_cli-0.1.6}/CHANGELOG.md

@@ -5,6 +5,25 @@ All notable changes to this project will be documented in this file.
 The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.1.0/),
 and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html).
 
+## [0.1.6] - 2026-05-09
+
+### Added
+- Add `aipa setup` interactive CLI command for first-run configuration (language, reference ticker, API key, base URL, model)
+- Add `setup_done` flag to user settings, auto-run setup before commands that need an API key (`analyze`, `deep-research`, TUI)
+- Add "First-Run Setup" section to README documenting which commands require setup
+
+### Changed
+- Lazy-load `AgentSession` in TUI instead of creating on mount, preventing crash when no API key is configured
+- Guard agent usage in chat (`_run_agent_chat`, `_run_analyze`, `/clear`) and settings tab with `_ensure_agent()`
+
+### Fixed
+- Fix crash on `aipa` launch when no API key is configured (`OpenAIError: Missing credentials`)
+
+## [0.1.5] - 2026-05-09
+
+### Changed
+- Expand README with Quick Start examples, full CLI reference for all commands, TUI slash commands, and configuration documentation
+
 ## [0.1.4] - 2026-05-09
 
 ### Added

aipa_cli-0.1.6/PKG-INFO

@@ -0,0 +1,260 @@
+Metadata-Version: 2.4
+Name: aipa-cli
+Version: 0.1.6
+Summary: Terminal TUI for AI-powered ticker analysis
+Project-URL: Homepage, https://github.com/quanhua92/aipriceaction
+Project-URL: Repository, https://github.com/quanhua92/aipriceaction
+Author-email: Quan Hua <quanhua92@gmail.com>
+License-Expression: MIT
+License-File: LICENSE
+Classifier: Development Status :: 3 - Alpha
+Classifier: Environment :: Console :: Curses
+Classifier: Intended Audience :: Financial and Insurance Industry
+Classifier: License :: OSI Approved :: MIT License
+Classifier: Programming Language :: Python :: 3
+Classifier: Programming Language :: Python :: 3.13
+Classifier: Topic :: Office/Business :: Financial
+Requires-Python: >=3.13
+Requires-Dist: aipriceaction>=0.1.9
+Requires-Dist: langchain-core
+Requires-Dist: langchain-openai
+Requires-Dist: langgraph
+Requires-Dist: textual-autocomplete>=4.0.6
+Requires-Dist: textual>=3.0.0
+Description-Content-Type: text/markdown
+
+# AIPA Terminal
+
+**Live site:** [aipriceaction.com](https://aipriceaction.com) | **GitHub:** [aipriceaction](https://github.com/quanhua92/aipriceaction) | **Frontend:** [aipriceaction-web](https://github.com/quanhua92/aipriceaction-web) | **Docker image:** [`quanhua92/aipriceaction:latest`](https://hub.docker.com/r/quanhua92/aipriceaction) | **Python SDK:** [`aipriceaction` on PyPI](https://pypi.org/project/aipriceaction/) | **AIPA Terminal:** [`aipa-cli` on PyPI](https://pypi.org/project/aipa-cli/)
+
+Textual-based terminal interface for AI-powered ticker analysis. Features streaming chat with thinking/reasoning display, autocomplete, slash commands, and workflow tabs.
+
+## Install
+
+```bash
+# Run directly (no install)
+uvx aipa-cli
+
+# Or install as a standalone tool
+uv tool install aipa-cli
+
+# Use either command
+aipa
+aipa-cli
+```
+
+## Requirements
+
+- Python 3.13+
+- An OpenAI-compatible API key (`OPENAI_API_KEY`) — only needed for AI analysis, not for data fetching
+- Optional: set `OPENAI_BASE_URL` for custom providers like OpenRouter
+
+## Quick Start
+
+```bash
+# Launch the TUI — first run auto-starts interactive setup
+aipa
+
+# Or run setup manually at any time
+aipa setup
+
+# AI analysis with default question template
+aipa analyze VCB
+
+# AI analysis with a custom question
+aipa analyze VCB --question "What is the support level and stop loss?"
+
+# Browse available question templates
+aipa analyze VCB --questions
+
+# Use a specific question template by index
+aipa analyze VCB 2 --question "What is the current trend of VCB?"
+
+# Dump raw context data without calling the LLM (no API key needed)
+aipa analyze VCB --context-only
+
+# Override language to English (default is your saved setting)
+aipa analyze VCB --lang en
+
+# Use hourly interval with SMA instead of EMA
+aipa analyze BTCUSDT --interval 1h --ma-type sma
+
+# Analyze multiple tickers at once
+aipa analyze VCB FPT VIC --interval 1D
+
+# Run multi-agent deep research pipeline
+aipa deep-research
+
+# Deep research with a custom question and save report to file
+aipa deep-research "Which sectors are leading the market?" --output report.md
+
+# Resume a previous deep-research session from checkpoint
+aipa deep-research --resume <session-id>
+
+# Fetch raw OHLCV data as a table
+aipa get-ohlcv-data VCB --interval 1D --limit 10
+
+# Fetch with date range and no moving averages
+aipa get-ohlcv-data VCB --start-date 2026-04-01 --end-date 2026-04-30 --no-ma
+```
+
+## CLI Commands
+
+### `aipa analyze`
+
+AI-powered analysis for one or more tickers. Builds context from OHLCV data and sends it to the LLM with a question.
+
+```
+# Default: uses question template 0 with your saved language setting
+aipa analyze VCB
+
+# Custom question
+aipa analyze VCB --question "Is this a good time to buy?"
+
+# List all question templates (trading opportunity, news, Wyckoff, etc.)
+aipa analyze VCB --questions
+
+# Raw data dump only (no LLM call, no API key required)
+aipa analyze VCB --context-only
+
+# Multi-ticker analysis
+aipa analyze VCB FPT MBB
+
+# Hourly data with SMA indicators
+aipa analyze VCB --interval 1h --ma-type sma --limit 50
+
+# Force English output
+aipa analyze VCB --lang en
+```
+
+| Flag | Description |
+|---|---|
+| `--question TEXT` | Custom analysis question |
+| `--questions` | List available question templates and exit |
+| `--context-only` | Dump raw context without LLM (no API key needed) |
+| `--interval` | Time interval: `1m`, `5m`, `15m`, `30m`, `1h`, `4h`, `1D`, `1W` (default: `1D`) |
+| `--limit N` | Number of bars (default: 20) |
+| `--source` | Filter by source: `vn` or `crypto` |
+| `--start-date` / `--end-date` | Date range (e.g. `2026-04-01`) |
+| `--reference-ticker` | Reference ticker for market context (default: `VNINDEX`) |
+| `--lang` | Language: `en` or `vn` (default: saved setting) |
+| `--ma-type` | Moving average type: `ema` or `sma` (default: `ema`) |
+
+### `aipa deep-research`
+
+Multi-agent deep research pipeline: supervisor decomposes into sector subtasks, parallel workers fetch data and analyze, aggregator synthesizes, and reviewer validates data integrity.
+
+```
+# Default: comprehensive market overview with all VN sectors
+aipa deep-research
+
+# Custom research question
+aipa deep-research "Compare banking vs real estate sectors"
+
+# Save final report to file
+aipa deep-research --output ~/reports/market-analysis.md
+
+# Resume from a previous checkpoint session
+aipa deep-research --resume 019e0cbb-0466-fa9f-d68c-2da40d35a68f
+
+# Force Vietnamese output
+aipa deep-research --lang vn
+```
+
+| Flag | Description |
+|---|---|
+| `--resume ID` | Resume from a checkpoint session ID |
+| `--output FILE` | Save final report to file |
+| `--lang` | Language: `en` or `vn` (default: saved setting) |
+
+### `aipa get-ohlcv-data`
+
+Fetch raw OHLCV data as a table (no LLM involved, works without setup).
+
+```
+# Default: daily data with EMA indicators
+aipa get-ohlcv-data VCB
+
+# Hourly data, last 10 bars
+aipa get-ohlcv-data VCB --interval 1h --limit 10
+
+# Date range, no moving averages
+aipa get-ohlcv-data VCB --start-date 2026-04-01 --end-date 2026-04-30 --no-ma
+
+# Crypto data
+aipa get-ohlcv-data BTCUSDT --interval 1D --limit 30
+```
+
+| Flag | Description |
+|---|---|
+| `--interval` | Time interval (default: `1D`) |
+| `--limit N` | Number of bars |
+| `--start-date` / `--end-date` | Date range |
+| `--source` | Filter by source: `vn` or `crypto` |
+| `--ma` / `--no-ma` | Include/exclude moving averages (default: included) |
+| `--ema` | Use EMA instead of SMA |
+
+### `aipa setup`
+
+Interactive first-run configuration. Prompts for language, reference ticker, API key, base URL, and model. Settings are saved to `~/.aipriceaction/settings.json`. Re-running shows current values as defaults.
+
+```
+# Run interactively
+aipa setup
+```
+
+## First-Run Setup
+
+Commands that require an API key will auto-run `aipa setup` on first use if not yet configured. Commands that don't need an API key always work immediately.
+
+| Command | Setup required? |
+|---|---|
+| `aipa get-ohlcv-data` | No setup needed |
+| `aipa analyze VCB --context-only` | No setup needed |
+| `aipa analyze VCB --questions` | No setup needed |
+| `aipa setup` | Runs setup |
+| `aipa` | Auto-runs setup first |
+| `aipa analyze VCB` | Auto-runs setup first |
+| `aipa deep-research` | Auto-runs setup first |
+
+## TUI
+
+Launch the TUI with `aipa`. The interface has three tabs:
+
+- **Chat** — AI-powered chat with streaming responses, thinking/reasoning display, slash commands, and arrow-key history navigation
+- **Workflows** — Structured analysis forms with question bank dropdown for ticker analysis and deep research
+- **Tickers** — Browse and search available tickers
+
+### Slash Commands (Chat tab)
+
+```
+/analyze VCB                  # Default AI analysis
+/analyze VCB 1h               # AI analysis with hourly interval
+/analyze VCB 2                # Use question template index 2
+/analyze VCB --question What is support?   # Custom question
+/export VCB FPT               # Export context to markdown file
+/deep-research                # Multi-agent research
+/clear                        # Clear chat history
+/exit                         # Quit
+```
+
+Press `Ctrl+O` in the Chat tab to view thinking/reasoning history.
+
+### Settings Tab
+
+Configure your API key, model, and base URL directly in the TUI. Settings are saved to `~/.aipriceaction/settings.json` and shared across both TUI and CLI.
+
+## Configuration
+
+Settings are loaded from `~/.aipriceaction/settings.json`. You can configure them via the TUI Settings tab or set environment variables:
+
+| Variable | Description | Default |
+|---|---|---|
+| `OPENAI_API_KEY` | API key for the LLM provider | — |
+| `OPENAI_BASE_URL` | Base URL for OpenAI-compatible API | OpenRouter |
+| `OPENAI_MODEL` | Model name | `openrouter/owl-alpha` |
+| `DATABASE_URL` | Backend API URL | `http://localhost:3000` |
+
+## License
+
+MIT

aipa_cli-0.1.6/README.md

@@ -0,0 +1,235 @@
+# AIPA Terminal
+
+**Live site:** [aipriceaction.com](https://aipriceaction.com) | **GitHub:** [aipriceaction](https://github.com/quanhua92/aipriceaction) | **Frontend:** [aipriceaction-web](https://github.com/quanhua92/aipriceaction-web) | **Docker image:** [`quanhua92/aipriceaction:latest`](https://hub.docker.com/r/quanhua92/aipriceaction) | **Python SDK:** [`aipriceaction` on PyPI](https://pypi.org/project/aipriceaction/) | **AIPA Terminal:** [`aipa-cli` on PyPI](https://pypi.org/project/aipa-cli/)
+
+Textual-based terminal interface for AI-powered ticker analysis. Features streaming chat with thinking/reasoning display, autocomplete, slash commands, and workflow tabs.
+
+## Install
+
+```bash
+# Run directly (no install)
+uvx aipa-cli
+
+# Or install as a standalone tool
+uv tool install aipa-cli
+
+# Use either command
+aipa
+aipa-cli
+```
+
+## Requirements
+
+- Python 3.13+
+- An OpenAI-compatible API key (`OPENAI_API_KEY`) — only needed for AI analysis, not for data fetching
+- Optional: set `OPENAI_BASE_URL` for custom providers like OpenRouter
+
+## Quick Start
+
+```bash
+# Launch the TUI — first run auto-starts interactive setup
+aipa
+
+# Or run setup manually at any time
+aipa setup
+
+# AI analysis with default question template
+aipa analyze VCB
+
+# AI analysis with a custom question
+aipa analyze VCB --question "What is the support level and stop loss?"
+
+# Browse available question templates
+aipa analyze VCB --questions
+
+# Use a specific question template by index
+aipa analyze VCB 2 --question "What is the current trend of VCB?"
+
+# Dump raw context data without calling the LLM (no API key needed)
+aipa analyze VCB --context-only
+
+# Override language to English (default is your saved setting)
+aipa analyze VCB --lang en
+
+# Use hourly interval with SMA instead of EMA
+aipa analyze BTCUSDT --interval 1h --ma-type sma
+
+# Analyze multiple tickers at once
+aipa analyze VCB FPT VIC --interval 1D
+
+# Run multi-agent deep research pipeline
+aipa deep-research
+
+# Deep research with a custom question and save report to file
+aipa deep-research "Which sectors are leading the market?" --output report.md
+
+# Resume a previous deep-research session from checkpoint
+aipa deep-research --resume <session-id>
+
+# Fetch raw OHLCV data as a table
+aipa get-ohlcv-data VCB --interval 1D --limit 10
+
+# Fetch with date range and no moving averages
+aipa get-ohlcv-data VCB --start-date 2026-04-01 --end-date 2026-04-30 --no-ma
+```
+
+## CLI Commands
+
+### `aipa analyze`
+
+AI-powered analysis for one or more tickers. Builds context from OHLCV data and sends it to the LLM with a question.
+
+```
+# Default: uses question template 0 with your saved language setting
+aipa analyze VCB
+
+# Custom question
+aipa analyze VCB --question "Is this a good time to buy?"
+
+# List all question templates (trading opportunity, news, Wyckoff, etc.)
+aipa analyze VCB --questions
+
+# Raw data dump only (no LLM call, no API key required)
+aipa analyze VCB --context-only
+
+# Multi-ticker analysis
+aipa analyze VCB FPT MBB
+
+# Hourly data with SMA indicators
+aipa analyze VCB --interval 1h --ma-type sma --limit 50
+
+# Force English output
+aipa analyze VCB --lang en
+```
+
+| Flag | Description |
+|---|---|
+| `--question TEXT` | Custom analysis question |
+| `--questions` | List available question templates and exit |
+| `--context-only` | Dump raw context without LLM (no API key needed) |
+| `--interval` | Time interval: `1m`, `5m`, `15m`, `30m`, `1h`, `4h`, `1D`, `1W` (default: `1D`) |
+| `--limit N` | Number of bars (default: 20) |
+| `--source` | Filter by source: `vn` or `crypto` |
+| `--start-date` / `--end-date` | Date range (e.g. `2026-04-01`) |
+| `--reference-ticker` | Reference ticker for market context (default: `VNINDEX`) |
+| `--lang` | Language: `en` or `vn` (default: saved setting) |
+| `--ma-type` | Moving average type: `ema` or `sma` (default: `ema`) |
+
+### `aipa deep-research`
+
+Multi-agent deep research pipeline: supervisor decomposes into sector subtasks, parallel workers fetch data and analyze, aggregator synthesizes, and reviewer validates data integrity.
+
+```
+# Default: comprehensive market overview with all VN sectors
+aipa deep-research
+
+# Custom research question
+aipa deep-research "Compare banking vs real estate sectors"
+
+# Save final report to file
+aipa deep-research --output ~/reports/market-analysis.md
+
+# Resume from a previous checkpoint session
+aipa deep-research --resume 019e0cbb-0466-fa9f-d68c-2da40d35a68f
+
+# Force Vietnamese output
+aipa deep-research --lang vn
+```
+
+| Flag | Description |
+|---|---|
+| `--resume ID` | Resume from a checkpoint session ID |
+| `--output FILE` | Save final report to file |
+| `--lang` | Language: `en` or `vn` (default: saved setting) |
+
+### `aipa get-ohlcv-data`
+
+Fetch raw OHLCV data as a table (no LLM involved, works without setup).
+
+```
+# Default: daily data with EMA indicators
+aipa get-ohlcv-data VCB
+
+# Hourly data, last 10 bars
+aipa get-ohlcv-data VCB --interval 1h --limit 10
+
+# Date range, no moving averages
+aipa get-ohlcv-data VCB --start-date 2026-04-01 --end-date 2026-04-30 --no-ma
+
+# Crypto data
+aipa get-ohlcv-data BTCUSDT --interval 1D --limit 30
+```
+
+| Flag | Description |
+|---|---|
+| `--interval` | Time interval (default: `1D`) |
+| `--limit N` | Number of bars |
+| `--start-date` / `--end-date` | Date range |
+| `--source` | Filter by source: `vn` or `crypto` |
+| `--ma` / `--no-ma` | Include/exclude moving averages (default: included) |
+| `--ema` | Use EMA instead of SMA |
+
+### `aipa setup`
+
+Interactive first-run configuration. Prompts for language, reference ticker, API key, base URL, and model. Settings are saved to `~/.aipriceaction/settings.json`. Re-running shows current values as defaults.
+
+```
+# Run interactively
+aipa setup
+```
+
+## First-Run Setup
+
+Commands that require an API key will auto-run `aipa setup` on first use if not yet configured. Commands that don't need an API key always work immediately.
+
+| Command | Setup required? |
+|---|---|
+| `aipa get-ohlcv-data` | No setup needed |
+| `aipa analyze VCB --context-only` | No setup needed |
+| `aipa analyze VCB --questions` | No setup needed |
+| `aipa setup` | Runs setup |
+| `aipa` | Auto-runs setup first |
+| `aipa analyze VCB` | Auto-runs setup first |
+| `aipa deep-research` | Auto-runs setup first |
+
+## TUI
+
+Launch the TUI with `aipa`. The interface has three tabs:
+
+- **Chat** — AI-powered chat with streaming responses, thinking/reasoning display, slash commands, and arrow-key history navigation
+- **Workflows** — Structured analysis forms with question bank dropdown for ticker analysis and deep research
+- **Tickers** — Browse and search available tickers
+
+### Slash Commands (Chat tab)
+
+```
+/analyze VCB                  # Default AI analysis
+/analyze VCB 1h               # AI analysis with hourly interval
+/analyze VCB 2                # Use question template index 2
+/analyze VCB --question What is support?   # Custom question
+/export VCB FPT               # Export context to markdown file
+/deep-research                # Multi-agent research
+/clear                        # Clear chat history
+/exit                         # Quit
+```
+
+Press `Ctrl+O` in the Chat tab to view thinking/reasoning history.
+
+### Settings Tab
+
+Configure your API key, model, and base URL directly in the TUI. Settings are saved to `~/.aipriceaction/settings.json` and shared across both TUI and CLI.
+
+## Configuration
+
+Settings are loaded from `~/.aipriceaction/settings.json`. You can configure them via the TUI Settings tab or set environment variables:
+
+| Variable | Description | Default |
+|---|---|---|
+| `OPENAI_API_KEY` | API key for the LLM provider | — |
+| `OPENAI_BASE_URL` | Base URL for OpenAI-compatible API | OpenRouter |
+| `OPENAI_MODEL` | Model name | `openrouter/owl-alpha` |
+| `DATABASE_URL` | Backend API URL | `http://localhost:3000` |
+
+## License
+
+MIT

{aipa_cli-0.1.4 → aipa_cli-0.1.6}/src/aipriceaction_terminal/__init__.py

@@ -1,4 +1,4 @@
 """AIPriceAction Terminal - TUI chat interface for ticker analysis."""
 
-__version__ = "0.1.4"
+__version__ = "0.1.6"
 

{aipa_cli-0.1.4 → aipa_cli-0.1.6}/src/aipriceaction_terminal/app.py

@@ -38,12 +38,12 @@ class AIPriceActionApp(AppActions, App):
         self.ticker = saved["ticker"]
         self.interval = saved["interval"]
         self.language = saved["language"]
+        self.agent = None
+        self._agent_lang: str | None = None
         from aipriceaction import AIContextBuilder
         from aipriceaction import AIPriceAction as AAPClient
         self.builder = AIContextBuilder(lang=self.language)
         self.client = AAPClient()
-        from .agents import AgentSession, AgentConfig
-        self.agent = AgentSession(AgentConfig(lang=self.language))
         self._load_ticker_options()
         # Populate SettingsTab widgets with loaded values
         self.query_one("#setting-ticker", Input).value = self.ticker
@@ -69,6 +69,19 @@ class AIPriceActionApp(AppActions, App):
         except Exception as e:
             self.notify(f"Failed to load tickers: {e}", severity="error")
 
+    def _ensure_agent(self) -> bool:
+        """Lazy-create AgentSession on first use. Returns False if API key is missing."""
+        if self.agent is not None and self._agent_lang == self.language:
+            return True
+        from aipriceaction.settings import settings
+        if not settings.openai_api_key:
+            self.agent = None
+            return False
+        from .agents import AgentSession, AgentConfig
+        self.agent = AgentSession(AgentConfig(lang=self.language))
+        self._agent_lang = self.language
+        return True
+
     def compose(self) -> ComposeResult:
         yield Header(show_clock=True)
         with TabbedContent(initial="chat"):

{aipa_cli-0.1.4 → aipa_cli-0.1.6}/src/aipriceaction_terminal/chat.py

@@ -216,7 +216,8 @@ class ChatTab(Vertical):
             )
         elif cmd == "/clear":
             log.clear()
-            self.app.agent.clear_history()
+            if self.app.agent is not None:
+                self.app.agent.clear_history()
         elif cmd == "/exit":
             self.app.exit()
         elif cmd == "/analyze":
@@ -322,6 +323,13 @@ class ChatTab(Vertical):
         """Build context and stream AI analysis for a ticker."""
         log = self.query_one("#chat-log", RichLog)
         try:
+            if not self.app._ensure_agent():
+                log.write(
+                    "[bold yellow]API key not configured.[/bold yellow]\n"
+                    "Set it in the Settings tab or run [bold]aipa setup[/bold]."
+                )
+                return
+
             builder = self.app.builder
 
             # Build context without system prompt (agent has it already)
@@ -389,6 +397,12 @@ class ChatTab(Vertical):
     async def _run_agent_chat(self, message: str) -> None:
         """Stream an agent response into the chat log."""
         log = self.query_one("#chat-log", RichLog)
+        if not self.app._ensure_agent():
+            log.write(
+                "[bold yellow]API key not configured.[/bold yellow]\n"
+                "Set it in the Settings tab or run [bold]aipa setup[/bold]."
+            )
+            return
         try:
             await stream_agent_to_log(
                 log,

{aipa_cli-0.1.4 → aipa_cli-0.1.6}/src/aipriceaction_terminal/cli.py

@@ -3,6 +3,14 @@
 import argparse
 
 
+def _ensure_setup() -> None:
+    """Run interactive setup if setup_done is not set."""
+    from .user_settings import load_settings
+    if not load_settings().get("setup_done"):
+        from .cli_setup import cmd_setup
+        cmd_setup()
+
+
 def run():
     parser = argparse.ArgumentParser(prog="aipa", description="AIPriceAction terminal")
     sub = parser.add_subparsers(dest="command")
@@ -44,15 +52,24 @@ def run():
     p_deep.add_argument("--output", default=None, help="Save final report to file")
     p_deep.add_argument("--lang", default=None, choices=["en", "vn"], help="Override language")
 
+    # aipa setup
+    sub.add_parser("setup", help="Interactive first-run setup")
+
     args = parser.parse_args()
 
-    if args.command == "analyze":
+    if args.command == "setup":
+        from .cli_setup import cmd_setup
+        cmd_setup()
+    elif args.command == "analyze":
+        if not getattr(args, "context_only", False) and not getattr(args, "questions", False):
+            _ensure_setup()
         from .cli_commands import cmd_analyze
         cmd_analyze(args)
     elif args.command == "get-ohlcv-data":
         from .cli_commands import cmd_get_ohlcv
         cmd_get_ohlcv(args)
     elif args.command == "deep-research":
+        _ensure_setup()
         from .cli_commands import cmd_deep_research
         cmd_deep_research(
             question=" ".join(args.question) if args.question else "",
@@ -61,5 +78,6 @@ def run():
             lang=args.lang,
         )
     else:
+        _ensure_setup()
         from .app import main
         main()

aipa_cli-0.1.6/src/aipriceaction_terminal/cli_setup.py

@@ -0,0 +1,71 @@
+"""Interactive first-run setup for aipa CLI (plain terminal, no TUI)."""
+
+from .user_settings import load_settings, save_settings
+
+
+_DEFAULT_BASE_URL = "https://openrouter.ai/api/v1"
+_DEFAULT_MODEL = "openai/gpt-oss-120b:free"
+
+
+def cmd_setup() -> None:
+    """Run interactive setup and save settings."""
+    current = load_settings()
+
+    print("AIPriceAction Terminal Setup")
+    print("=" * 30)
+
+    # Language
+    lang = _prompt(
+        "Language (en/vn)",
+        current.get("language", "en"),
+    )
+    if lang not in ("en", "vn"):
+        print("Invalid language, defaulting to 'en'.")
+        lang = "en"
+
+    # Reference ticker
+    ticker = _prompt(
+        "Reference ticker",
+        current.get("ticker", "VNINDEX"),
+    )
+
+    # API key (optional)
+    api_key = _prompt(
+        "API key (press Enter to skip)",
+        current.get("api_key", ""),
+    )
+
+    # Base URL
+    base_url = _prompt(
+        "Base URL",
+        current.get("openai_base_url") or _DEFAULT_BASE_URL,
+    )
+
+    # Model
+    model = _prompt(
+        "Model",
+        current.get("openai_model") or _DEFAULT_MODEL,
+    )
+
+    data = {
+        "ticker": ticker.upper(),
+        "interval": current.get("interval", "1D"),
+        "language": lang,
+        "api_key": api_key,
+        "openai_base_url": base_url,
+        "openai_model": model,
+        "setup_done": True,
+    }
+
+    save_settings(data)
+    print("Setup complete.")
+
+
+def _prompt(label: str, default: str) -> str:
+    """Prompt user with current default shown in brackets."""
+    hint = f" [{default}]" if default else ""
+    try:
+        value = input(f"{label}{hint}: ").strip()
+    except EOFError:
+        return default
+    return value if value else default

{aipa_cli-0.1.4 → aipa_cli-0.1.6}/src/aipriceaction_terminal/settings_tab.py

@@ -151,8 +151,13 @@ class SettingsTab(Vertical):
                 self.app.language = language
                 from aipriceaction import AIContextBuilder
                 self.app.builder = AIContextBuilder(lang=language)
-                from .agents import AgentSession, AgentConfig
-                self.app.agent = AgentSession(AgentConfig(lang=language))
+                try:
+                    from .agents import AgentSession, AgentConfig
+                    self.app.agent = AgentSession(AgentConfig(lang=language))
+                    self.app._agent_lang = language
+                except Exception:
+                    self.app.agent = None
+                    self.app._agent_lang = None
 
             data: dict = {
                 "ticker": ticker,

{aipa_cli-0.1.4 → aipa_cli-0.1.6}/src/aipriceaction_terminal/user_settings.py

@@ -13,6 +13,7 @@ _DEFAULTS = {
     "api_key": "",
     "openai_base_url": "",
     "openai_model": "",
+    "setup_done": False,
 }
 
 

aipa_cli-0.1.4/PKG-INFO

@@ -1,73 +0,0 @@
-Metadata-Version: 2.4
-Name: aipa-cli
-Version: 0.1.4
-Summary: Terminal TUI for AI-powered ticker analysis
-Project-URL: Homepage, https://github.com/quanhua92/aipriceaction
-Project-URL: Repository, https://github.com/quanhua92/aipriceaction
-Author-email: Quan Hua <quanhua92@gmail.com>
-License-Expression: MIT
-License-File: LICENSE
-Classifier: Development Status :: 3 - Alpha
-Classifier: Environment :: Console :: Curses
-Classifier: Intended Audience :: Financial and Insurance Industry
-Classifier: License :: OSI Approved :: MIT License
-Classifier: Programming Language :: Python :: 3
-Classifier: Programming Language :: Python :: 3.13
-Classifier: Topic :: Office/Business :: Financial
-Requires-Python: >=3.13
-Requires-Dist: aipriceaction>=0.1.9
-Requires-Dist: langchain-core
-Requires-Dist: langchain-openai
-Requires-Dist: langgraph
-Requires-Dist: textual-autocomplete>=4.0.6
-Requires-Dist: textual>=3.0.0
-Description-Content-Type: text/markdown
-
-# AIPA Terminal
-
-**Live site:** [aipriceaction.com](https://aipriceaction.com) | **GitHub:** [aipriceaction](https://github.com/quanhua92/aipriceaction) | **Frontend:** [aipriceaction-web](https://github.com/quanhua92/aipriceaction-web) | **Docker image:** [`quanhua92/aipriceaction:latest`](https://hub.docker.com/r/quanhua92/aipriceaction) | **Python SDK:** [`aipriceaction` on PyPI](https://pypi.org/project/aipriceaction/) | **AIPA Terminal:** [`aipa-cli` on PyPI](https://pypi.org/project/aipa-cli/)
-
-Textual-based terminal interface for AI-powered ticker analysis. Features streaming chat with thinking/reasoning display, autocomplete, slash commands, and workflow tabs.
-
-## Install
-
-```bash
-# Run directly (no install)
-uvx aipa-cli
-
-# Or install as a standalone tool
-uv tool install aipa-cli
-
-# Use either command
-aipa
-aipa-cli
-```
-
-## Requirements
-
-- Python 3.13+
-- An OpenAI-compatible API key (`OPENAI_API_KEY`)
-- Optional: set `OPENAI_BASE_URL` for custom providers like OpenRouter
-
-## Usage
-
-```
-aipa              # Launch the TUI
-aipa analyze      # Run ticker analysis from CLI
-aipa get-ohlcv-data  # Fetch OHLCV data from CLI
-aipa deep-research   # Run deep research from CLI
-```
-
-### TUI
-
-The interface has three tabs:
-
-- **Chat** — AI-powered chat with streaming responses, thinking/reasoning display, slash commands (`/analyze`, `/export`, `/clear`, `/exit`), and arrow-key history navigation
-- **Workflows** — Structured analysis forms for ticker analysis and deep research
-- **Tickers** — Browse and search available tickers
-
-Press `Ctrl+O` in the Chat tab to view thinking/reasoning history.
-
-## License
-
-MIT

aipa_cli-0.1.4/README.md

@@ -1,48 +0,0 @@
-# AIPA Terminal
-
-**Live site:** [aipriceaction.com](https://aipriceaction.com) | **GitHub:** [aipriceaction](https://github.com/quanhua92/aipriceaction) | **Frontend:** [aipriceaction-web](https://github.com/quanhua92/aipriceaction-web) | **Docker image:** [`quanhua92/aipriceaction:latest`](https://hub.docker.com/r/quanhua92/aipriceaction) | **Python SDK:** [`aipriceaction` on PyPI](https://pypi.org/project/aipriceaction/) | **AIPA Terminal:** [`aipa-cli` on PyPI](https://pypi.org/project/aipa-cli/)
-
-Textual-based terminal interface for AI-powered ticker analysis. Features streaming chat with thinking/reasoning display, autocomplete, slash commands, and workflow tabs.
-
-## Install
-
-```bash
-# Run directly (no install)
-uvx aipa-cli
-
-# Or install as a standalone tool
-uv tool install aipa-cli
-
-# Use either command
-aipa
-aipa-cli
-```
-
-## Requirements
-
-- Python 3.13+
-- An OpenAI-compatible API key (`OPENAI_API_KEY`)
-- Optional: set `OPENAI_BASE_URL` for custom providers like OpenRouter
-
-## Usage
-
-```
-aipa              # Launch the TUI
-aipa analyze      # Run ticker analysis from CLI
-aipa get-ohlcv-data  # Fetch OHLCV data from CLI
-aipa deep-research   # Run deep research from CLI
-```
-
-### TUI
-
-The interface has three tabs:
-
-- **Chat** — AI-powered chat with streaming responses, thinking/reasoning display, slash commands (`/analyze`, `/export`, `/clear`, `/exit`), and arrow-key history navigation
-- **Workflows** — Structured analysis forms for ticker analysis and deep research
-- **Tickers** — Browse and search available tickers
-
-Press `Ctrl+O` in the Chat tab to view thinking/reasoning history.
-
-## License
-
-MIT