Python SDK + AI Agent for Sentinel — 80+ crypto trading, AI, and market intelligence tools with interactive chat. Free access, usage-based fees.
Project description
Local-First AI Agent for Crypto, Equities & Macro
12 data sources · 65+ tools · Multi-agent swarm · Sub-second queries · One pip install
Website · Docs · API Reference · PyPI
Install
pip install hyper-sentinel && sentinel-chat
Paste your LLM key on first run. No email, no account. You're in.
With multi-agent swarm (Upsonic):
pip install 'hyper-sentinel[swarm]' && sentinel-chat
How It Works
Sentinel is an AI agent that calls tools to answer your questions. Common queries hit a fast path and return instantly. Everything else goes through your chosen LLM.
⚡ You → price of btc and xmr
╭─────────────────────── 🛡️ Sentinel ──────────────────────────╮
│ │
│ Bitcoin (BTC): $66,565.00 │
│ 24h: -0.70% · 7d: -3.28% · Market cap: $1.3T │
│ │
│ Monero (XMR): $330.89 │
│ 24h: +0.18% · 7d: -6.13% · Market cap: $6.1B │
│ │
╰────────────────────────────────────── ⚡ instant · 0 LLM ─────╯
| Query Type | Speed | How |
|---|---|---|
price of btc / top 10 crypto |
< 1 second | Fast Path — regex match, direct API call, zero LLM |
analyze BTC structure and AAPL recs |
3–8 seconds | LLM agent calls tools, formats response |
run quant analysis on TSLA |
10–20 seconds | Deep quant engine — valuation, technicals, risk, analyst targets |
Multi-Agent Swarm
Sentinel includes a Upsonic-powered multi-agent swarm for coordinated analysis and trading. Three specialized agents work together:
| Agent | Role | Capabilities |
|---|---|---|
| 📊 Analyst | Market Research & Macro | Crypto prices, stock analysis, FRED data, sentiment, news |
| ⚠️ RiskManager | Portfolio Risk & Sizing | Position limits, leverage checks, trade approval/rejection |
| 💰 Trader | Trade Execution | Hyperliquid, Aster DEX, Polymarket order placement |
# Install with swarm support
pip install 'hyper-sentinel[swarm]'
⚡ You → swarm
🤖 Initializing Sentinel Swarm...
🛡️ Sentinel Swarm — ONLINE
Analyst ● ONLINE sentinel.analyst
RiskManager ● ONLINE sentinel.risk
Trader ● ONLINE sentinel.trader
3 agents · Mode: COORDINATE · 42 tools · anthropic/claude-sonnet-4-20250514
⚡ You → analyze BTC macro outlook and assess trade risk
╭─────────────────────── 🛡️ Sentinel Swarm ────────────────────╮
│ [Analyst researches → RiskManager evaluates → coordinated] │
╰──────────────────────────────── 3 agents · coordinate · 14s ──╯
Python API:
from sentinel.swarm import build_swarm, swarm_chat
team, agents = build_swarm(mode="coordinate")
result = swarm_chat(team, "analyze BTC macro outlook and assess trade risk")
Data Sources
All 12 sources run locally via built-in scrapers. No gateway, no cloud dependency.
| Source | Tools | Setup |
|---|---|---|
| 🪙 CoinGecko | get_crypto_price · get_crypto_top · search_crypto |
None — always available |
| 📈 YFinance | get_stock_quote · run_stock_analysis · get_analyst_recs |
None — always available |
| 📊 DexScreener | dexscreener_search · dexscreener_trending |
None — always available |
| 🏛️ FRED | get_fred_series · search_fred · get_economic_dashboard |
add fred |
| 📰 Y2 Intelligence | get_news_sentiment · get_news_recap · get_intelligence_reports |
add y2 |
| 🔮 Elfa AI | get_trending_tokens · get_top_mentions · search_mentions |
add elfa |
| 🐦 X / Twitter | search_x |
add x |
| ⚡ Hyperliquid | get_hl_positions · place_hl_order · get_hl_orderbook + 7 more |
add hl |
| 🌟 Aster DEX | aster_ticker · aster_positions · aster_balance + 7 more |
add aster |
| 🎲 Polymarket | get_polymarket_markets · buy_polymarket + 8 more |
add polymarket |
| 💬 Telegram | tg_read_channel · tg_search_messages · tg_list_channels |
add telegram |
| 🎮 Discord | discord_read_channel · discord_list_guilds + 4 more |
add discord |
3 sources work out of the box. The rest activate the moment you add a key — no restart needed.
Python SDK
from sentinel import SentinelClient
client = SentinelClient() # keys loaded from ~/.sentinel/config
# Crypto (instant)
btc = client.get_crypto_price("bitcoin")
top = client.get_crypto_top_n(10)
# Equities — deep quant analysis
aapl = client.get_stock_price("AAPL")
analysis = client.run_stock_analysis("TSLA") # valuation + technicals + risk
# Perpetual futures (Hyperliquid)
positions = client.get_hl_positions()
client.place_hl_order(coin="ETH", side="buy", size=0.5)
# Prediction markets (Polymarket)
markets = client.get_polymarket_markets()
# Macro data (FRED)
gdp = client.get_fred_series("GDP")
# AI chat — uses your LLM
resp = client.chat("What's the best DeFi play right now?")
Authentication
Add your LLM API key. That's it.
sentinel-chat # prompts on first run
sentinel-setup # or run the setup wizard
client = SentinelClient(ai_key="sk-ant-xxx") # or pass directly
| Provider | Prefix | Get a key |
|---|---|---|
| Anthropic (Claude) | sk-ant- |
console.anthropic.com |
| OpenAI (GPT) | sk- |
platform.openai.com |
| Google (Gemini) | AIza |
aistudio.google.com |
| xAI (Grok) | xai- |
console.x.ai |
Commands
Chat (inside sentinel-chat)
add List all data sources you can configure
add hl Hyperliquid perp futures
add polymarket Polymarket prediction markets
add aster Aster DEX futures
add fred FRED economic data
add x X / Twitter
add y2 Y2 Intelligence news
add elfa Elfa AI social intel
add telegram Telegram channels
add discord Discord bot
add ai Switch LLM provider
swarm Start multi-agent mode (Analyst + Risk + Trader)
swarm status Show agent roster and status
solo Return to single-agent mode
tools List all 65+ tools
status Infrastructure dashboard
clear Reset context
quit Exit
Terminal
sentinel-chat # Interactive agent
sentinel-setup # Onboarding wizard
sentinel ask "price of ETH" # One-shot query
sentinel status # Dashboard
sentinel wallet # Manage wallets
sentinel tools # List all tools
Architecture
sentinel-chat / sentinel ask / SentinelClient
│
┌────────────┼────────────┐
│ │ │
Fast Path 12 Local Upsonic
(regex) Scrapers Swarm
│ │ │
< 1 sec CoinGecko 3 Agents
zero LLM YFinance Analyst
DexScreener RiskManager
FRED Trader
Y2 · Elfa │
X · HL coordinate
Aster · PM mode
TG · Discord
- Fast Path → regex-matched price/top queries → instant
- Local Scrapers → all tool calls execute locally via 12 built-in modules
- Swarm → optional Upsonic multi-agent coordination (install with
[swarm])
Error Handling
from sentinel.exceptions import SentinelError, AuthError, RateLimitError
try:
client.place_hl_order(coin="ETH", side="buy", size=0.1)
except RateLimitError as e:
print(f"Rate limited — retry after {e.retry_after}s")
except AuthError:
print("Invalid credentials")
Changelog
v0.3.5 — Config Bridge + Bug Fixes
- FIX: Bridge
~/.sentinel/configcredentials to env vars for all 14 scraper keys - FIX: Aster secret and Polymarket funder env var mismatches
v0.3.4 — Multi-Agent Swarm
- ADD: Upsonic multi-agent swarm (Analyst, RiskManager, Trader)
- ADD:
swarm,swarm status,soloREPL commands - ADD: Dashboard shows swarm availability and agent roster
- ADD: SQLite memory persistence for swarm sessions
v0.3.3 — Quant Analysis Engine
- ADD:
run_stock_analysis— deep quantitative analysis (valuation, technicals, risk, analyst targets) - UPGRADE: System prompt with 11-section analysis formatting guide
v0.3.2 — Stabilization
- FIX: Tool-naming mismatches between chat module and scrapers
- ADD: YFinance local-first stock data integration
Links
- hyper-sentinel.com — Website
- Docs — Full documentation
- API Reference — Every endpoint
- PyPI — Package
- Upsonic — Multi-agent framework
License
AGPL-3.0 — © 2026 Sentinel Labs
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
Built Distribution
Filter files by name, interpreter, ABI, and platform.
If you're not sure about the file name format, learn more about wheel file names.
Copy a direct link to the current filters
File details
Details for the file hyper_sentinel-0.3.8.tar.gz.
File metadata
- Download URL: hyper_sentinel-0.3.8.tar.gz
- Upload date:
- Size: 103.1 kB
- Tags: Source
- Uploaded using Trusted Publishing? No
- Uploaded via: twine/6.2.0 CPython/3.12.0
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 |
7112def3bda0adc5092771f527033ea0b82c4eb959891a64ec853fa0b2fc54e0
|
|
| MD5 |
cc0c511baf9bb43e6114851f3a29722a
|
|
| BLAKE2b-256 |
eff5498e8ee34427499efad358bd6e38e887b120592a35e8b45373a2db1caf92
|
File details
Details for the file hyper_sentinel-0.3.8-py3-none-any.whl.
File metadata
- Download URL: hyper_sentinel-0.3.8-py3-none-any.whl
- Upload date:
- Size: 109.3 kB
- Tags: Python 3
- Uploaded using Trusted Publishing? No
- Uploaded via: twine/6.2.0 CPython/3.12.0
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 |
dd59ff7dd3a51db961e366f163d116bd43434afcb6525b51617c87e44309b112
|
|
| MD5 |
045fe7f5fcbec4d70d4f081ed70f2643
|
|
| BLAKE2b-256 |
a60f8fe56569f28ac3fff590b46a568994fb4d549db28393089bfe6100496e91
|