sigmavue-mcp
MCP server for controlling SigmaVue trading accounts from the terminal or MCP-speaking assistants. Provides tools for account status, journal, portfolio, simulated trade execution, and copy-trading group management.
README
SigmaVue CLI
Control your SigmaVue trading account from the terminal —
and from any MCP-speaking assistant. sigmavue is a thin, read-mostly client
over the public SigmaVue API: check status, journal, and portfolio; manage broker
accounts and copy-trading groups; and place kill-switch-gated simulated trades —
without clicking through the app.
It adds no backend logic and bundles no trading strategies. Every command maps to an API endpoint that already ships. Trade execution is fail-closed and SIM-only.
Install
Run it without installing (requires uv):
uvx --from git+https://github.com/Sigma-Algo/sigmavue-cli sigmavue --help
Install from source:
git clone https://github.com/Sigma-Algo/sigmavue-cli
cd sigmavue-cli
uv sync # or: pip install -e .
uv run sigmavue --help
pip install sigmavue— coming soon (PyPI publish pending).
Quickstart
# Point at the API (optional). Defaults to the public SigmaVue API.
export SIGMAVUE_API_URL=https://sigmavue-api-green.onrender.com
# or http://localhost:8000 for a local backend.
sigmavue login --email you@example.com # password read at a secure prompt
sigmavue status # health + account summary
sigmavue accounts list # broker accounts your token owns
sigmavue journal list --symbol NQ # trade journal
sigmavue portfolio # open positions + rollup
sigmavue trade execute --symbol NQ --qty 1 --side buy --sim # SIM-only
sigmavue copy groups list # copy-trading groups
The bearer token is stored at ~/.sigmavue/auth.json (file mode 0600).
Passwords are only ever read from a prompt, never accepted as an argument.
Commands
| Command | What it does |
|---|---|
login / logout |
Authenticate; store or clear the local token |
status |
Backend health + current user + account summary |
accounts list |
List broker accounts your token is scoped to (--active) |
accounts add |
Start a broker OAuth connect (returns an auth_url to open) |
accounts remove |
Remove a broker account by id |
clients list |
List managed trading profiles (enterprise) |
journal list |
List trade-journal entries (--symbol, --type, paging) |
portfolio |
Current positions + summary (--account, --status) |
trade execute |
Place an order — kill-switch gated, --sim required |
copy groups list |
List copy-trading groups |
copy groups create |
Create a copy-trading group |
copy groups add-follower |
Coming soon — use the app for now (see below) |
Exit codes: 0 success · 1 validation / API / trade-refused · 2 not logged in.
Trading safety
Trade execution is fail-closed and re-checked on every call, before any order request leaves your machine:
-
CLI live-execution is disabled —
--simis required; SIM routes to your active demo broker account. -
Local kill-switch — the CLI refuses to place an order whenever trading is turned off, via any of:
- env
TRADING_ENABLEDset to0/false/no/off ~/.sigmavue/killswitch.jsonwith{"paused": true}(a missing or corrupt file also blocks — fail-closed)~/.sigmavue/TRADING_ENABLEDfile containing0
Paths are overridable via
SIGMAVUE_KILLSWITCH_PATHandSIGMAVUE_TRADING_ENABLED_PATH. - env
-
Backend circuit breaker — the order is refused unless the server-side per-user circuit breaker is
CLOSED.
Rate limits are respected: a stable User-Agent/X-SigmaVue-Client is sent, and
429 responses trigger bounded Retry-After backoff (--verbose surfaces the
rate-limit headers).
MCP server
The same capability layer is exposed as an MCP
server, so phone / agent coding assistants get identical reach over one API. It
speaks stdio and reads the SAME token sigmavue login stores — there is no
credential collection in the MCP surface.
uv sync --extra mcp # or: pip install "sigmavue[mcp]"
sigmavue login # once, in a terminal
uv run sigmavue-mcp # serves over stdio
Add it to an MCP client (use an absolute path so it can start from anywhere):
{
"mcpServers": {
"sigmavue": {
"command": "uv",
"args": [
"--directory", "/ABSOLUTE/PATH/TO/sigmavue-cli",
"run", "sigmavue-mcp"
]
}
}
}
Claude Code:
claude mcp add sigmavue -- \
uv --directory /ABSOLUTE/PATH/TO/sigmavue-cli run sigmavue-mcp
It exposes eight tools: sigmavue_status, sigmavue_accounts_list,
sigmavue_clients_list, sigmavue_journal_list, sigmavue_portfolio,
sigmavue_trade_execute, sigmavue_copy_groups_list, and
sigmavue_copy_groups_create. Interactive/destructive flows (login, broker
connect/remove) stay CLI-only.
Enterprise
Command breadth is decided server-side by your token: a member sees their own
accounts, a manager/enterprise token sees everything it owns — the same
accounts list command covers both. A dedicated multi-tenant clients endpoint is
on the roadmap; clients list maps to managed trading profiles today.
Adding copy-trade followers (coming soon)
copy groups add-follower is intentionally not enabled from the CLI or MCP yet:
there is no atomic add-follower endpoint, and a client-side read-modify-write of
the follower list would be unsafe. A server-side
POST /api/v2/trading-groups/{id}/followers is planned; until then, add
followers in the SigmaVue app. copy groups list and create work fully.
Development
uv sync
uv run pytest -q # tests use httpx.MockTransport — no live API calls
License
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.