llama-index-tools-alphai

LlamaIndex tool spec for AlphaAI — AI-scored, ticker-linked financial news and SEC Form 4 insider events, built for AI agents and trading bots.

Every article on the AlphaAI feed is enriched before you see it: per-ticker impact analysis, one of 14 categories, and a 1-10 market-relevance score. The tools here fetch and filter that feed — no scraping, no scoring of your own.

AlphaAIToolSpec exposes four tools (each with a native async twin):

• search_news — the scored news feed with ticker / category / relevance filters
• get_insider_news — SEC Form 4 insider events with a structured who-sold-what block
• get_ticker_sentiment — 7-day bullish/neutral/bearish rollup for one ticker
• get_insider_summary — 30-day insider buy/sell rollup for one ticker

Install

pip install llama-index-tools-alphai

Requires Python 3.10+.

Authentication

Create an API key at alphai.io/developers — the free tier works without a card. Export it as ALPHAI_API_KEY, or pass api_key=... to the spec.

export ALPHAI_API_KEY="ak_live_..."

Use in an agent

from llama_index.core.agent.workflow import FunctionAgent
from llama_index.llms.anthropic import Anthropic

from llama_index.tools.alphai import AlphaAIToolSpec

agent = FunctionAgent(
    tools=AlphaAIToolSpec().to_tool_list(),
    llm=Anthropic(model="claude-sonnet-4-5"),
)

response = await agent.run(
    "Are NVDA insiders net buying or selling this month, "
    "and does news sentiment agree with what they're doing?"
)

Use the tools directly

from llama_index.tools.alphai import AlphaAIToolSpec

spec = AlphaAIToolSpec()

docs = spec.search_news(symbol="NVDA", min_relevance=7, max_results=5)
docs[0].text  # title + summary
docs[0].metadata  # tickers, category, relevance_score, url, source, ...

events = spec.get_insider_news(symbol="NVDA", min_relevance=7)
events[0].metadata["insider"]  # side, shares, avg price, total value, 10b5-1 flag

spec.get_ticker_sentiment("NVDA")  # {"bullish": 30, "neutral": 8, "bearish": 4, ...}
spec.get_insider_summary("NVDA")  # 30-day buy/sell counts and dollar values

News documents map cleanly into RAG pipelines — text is the article title plus summary, everything else rides in metadata.

Filters

• symbol — one ticker ("NVDA"; crypto "BTC-USD", foreign listings the Yahoo suffix "VOD.L"). Share-class siblings are included on the insider feed.
• category — one of 14: earnings, mergers_acquisitions, regulation, macro_economy, sector_analysis, market_movers, technology, commodities, crypto, ipo, geopolitics, insider, corporate_actions, other.
• min_relevance — 1-10 floor; 7+ keeps only high-signal articles. On the insider feed the score is deterministic from the event's dollar value, so it works as an "only large trades" dial.
• max_results — cap on returned items (1-50).

Rate limits

Limits are per AlphaAI account, two-layer (per-minute burst + per-day volume): Free 20/min · 100/day, Basic 60/min · 10,000/day, Pro 150/min · 100,000/day. The underlying alphai-sdk retries 429s with backoff automatically.

Links

• Developer guide: https://alphai.io/developers
• API reference: https://api.alphai.io/api/schema/
• MCP server (same feed, for MCP-speaking agents): https://alphai.io/mcp

AlphaAI output is AI-generated financial information for research, not investment advice — see alphai.io/terms.