aipa-cli 0.1.13

Terminal TUI for AI-powered ticker analysis

AIPA Terminal

Live site: aipriceaction.com | GitHub: aipriceaction | Frontend: aipriceaction-web | Docker image: quanhua92/aipriceaction:latest | Python SDK: aipriceaction on PyPI | AIPA Terminal: aipa-cli on PyPI

Textual-based terminal interface for AI-powered ticker analysis. Features streaming chat with thinking/reasoning display, autocomplete, slash commands, and workflow tabs.

Install

# 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

# 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

# List saved chat sessions
aipa resume

# Open TUI with a resumed session
aipa resume 0

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

aipa resume

List saved chat sessions or open the TUI with a resumed one. Sessions are auto-saved under ~/.aipriceaction/sessions/.

# List recent sessions
aipa resume

# Open TUI with a specific session
aipa resume 0

# Open TUI by UUID prefix
aipa resume 0194a2b3

Argument	Description
session	Session ID prefix or list index (omit to list all)

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 resume	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 six tabs:

• Chat — AI-powered chat with streaming responses, thinking/reasoning display, slash commands, and arrow-key history navigation
• Vietnam — Browse and search Vietnamese stock tickers
• Crypto — Browse and search cryptocurrency tickers
• Global — Browse and search global/Yahoo tickers
• Workflows — Structured analysis forms with question bank dropdown for ticker analysis and deep research
• Settings — Configure API key, model, base URL, and other preferences

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
/save                         # Export chat to markdown (default: ~/aipriceaction-chat-<id>.md)
/save ~/my-chat.md            # Export chat to custom path
/resume                       # List saved sessions
/resume 0                     # Load session by index
/resume 0194a2b3              # Load session by UUID prefix
/sessions                     # Alias for /resume
/new                          # Start new session (clears history)
/clear                        # Clear chat display only
/exit                         # Quit

Chat sessions are auto-saved to ~/.aipriceaction/sessions/ and restored with full LLM context when resumed.

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