Skip to main content
Glama

Otto

CI

A local financial terminal your AI agent operates for you. Market data, backtests, paper books, news — driven in plain language, running entirely on your machine.

A real agent operating Otto: a plain-language command goes in, the agent runs a backtest through MCP, and the run lands on the dashboard

Try it

Needs uv, Python 3.12 and Claude Code. Install once, then register it:

uv tool install git+https://github.com/0Smallcat0/otto
claude mcp add otto -- otto-terminal

Install first, in a terminal: on a slow connection the first download can outlast the 30 or so seconds Claude Code gives an MCP server to start. If uv says its bin directory is not on your PATH, run uv tool update-shell and open a new terminal. Update later with uv tool upgrade otto-terminal.

Then ask for things:

  • "Refresh market data and show me the BTC snapshot."

  • "Backtest an SMA cross on BTCUSDT and tell me if it's any good."

  • "Use my paper book to buy $5,000 of AAPL, say why in one line, then show me the book."

The dashboard comes with it. The MCP server starts the backend for you, and http://127.0.0.1:8765/ is the screen: books, calls and every order the agent placed.

Related MCP server: OpenFinClaw CLI

Run a round every day

A round is what the terminal is for: grade the open calls, scan for names with no view, record a call only where there is a real thesis, place paper orders with a reason, then snapshot the books. To run one, ask:

"Run one Otto decision round: grade my open calls, scan for gaps, record a call only where you have a real thesis, place paper orders with a rationale, then snapshot the books and report each against its benchmark."

To have it run every morning on its own (Windows, PowerShell), save the runner and register it once:

iwr https://raw.githubusercontent.com/0Smallcat0/otto/main/scripts/run_decision_loop.ps1 -OutFile "$HOME\otto-round.ps1"
$run = New-ScheduledTaskAction -Execute powershell -Argument "-NoProfile -ExecutionPolicy Bypass -File `"$HOME\otto-round.ps1`""
Register-ScheduledTask OttoDailyRound -Action $run -Trigger (New-ScheduledTaskTrigger -Daily -At 9am)

Each run is one headless claude -p session on your Claude account (a test round on 2026-10-22 took 3 minutes, 33 turns and $1.18). It writes the theses in your Windows display language — add -Language English after the .ps1 path to pick another — and appends its duration, turns and cost to ~/.otto/artifacts/loop/runs.jsonl.

Why this one

It keeps score. Every call the agent records — a stance, the reasoning, the price that would prove it wrong, a horizon — is graded later against the price the market actually printed and against its index (0050.TW, SPY, BTC-USD); a call without reasoning is refused. "An AI can operate it" is measured: a real headless agent runs plain-language tasks in hermetic sandboxes, graded on terminal state and artifacts, never by an LLM judge.

Model

Passed

Avg turns

Run

claude-sonnet-5

21 / 21

7.7

2026-08-16, 21-task suite

claude-haiku-4-5

21 / 21

6.4

2026-08-16, 21-task suite

Live trading, credential entry and code execution aren't switched off — they're unreachable through the surface the agent has.

Under the hood

One typed contract (142 actions across 16 routes) is the single source of truth; the MCP tools, the UI capability catalog and the eval suite are all derived from it. 805 tests on Windows + Linux CI.

The tests and the eval harness are not in the installed package, so these run from a clone. The built dashboard is committed, so uv run otto needs no frontend build:

git clone https://github.com/0Smallcat0/otto && cd otto
uv run otto
uv run --extra dev python -m pytest -q
python evals/run_eval.py --model claude-sonnet-5

License

MIT.

Available Tools

6 tools
get_routeA

Fetch the current state of one route (e.g. dashboard, markets, portfolio, backtest, command center is under settings). Returns the route's primary read-only payload.

Inputs: route_id (required).

list_actionsA

List the safe operable actions (from the terminal agent-contract). Excludes safety-disabled and secret actions. Each row shows action_id, route_id, label, method, endpoint, request_contract, safety_class, whether it mutates local state, and whether it needs confirmation. Use action_id with run_action.

Inputs: route_id.

list_routesA

List the 15 terminal routes with their route_id, label, category, and primary endpoint. Use route_id with get_route.

Inputs: no named parameters.

refresh_public_dataA

Convenience: start the public no-key provider refresh job and return its result summary (written / available / reused cache counts). This is the common 'load fresh public data' operation; no keys or private accounts.

Inputs: no named parameters.

run_actionA

Execute one safe terminal action by action_id (from list_actions). Refuses safety-disabled and secret actions. For POST actions pass 'body'; for endpoints with {placeholders} pass 'path_params'; optional 'query'. If the action requires confirmation, pass confirm=true.

Inputs: action_id (required), body, query, confirm, +1 more.

terminal_statusA

Situational awareness entry point. Returns terminal health plus the Command Center summary: current milestone, goal status, active task, risk gates (live/secrets), recovery count, and provider freshness. Call this first before operating.

Inputs: no named parameters.

Tool Schema Changelog

Recent tool additions, removals, and schema changes observed during successful MCP inspections.

Loading tool schema changelog…

    TDQS

    A4.3/5.0

    Scored across 6 tools

    Disambiguation5/5

    Each tool has a clearly distinct purpose: fetching a route, listing actions, listing routes, refreshing data, executing an action, and checking terminal status. No overlap or ambiguity.

    Naming Consistency5/5

    All tool names follow a consistent verb_noun pattern with snake_case (e.g., get_route, list_actions, run_action). This makes them predictable and easy to understand.

    Tool Count5/5

    Six tools is well-scoped for a terminal interface. Each tool serves a necessary function without redundancy, covering status, navigation, data refresh, and action execution.

    Completeness4/5

    The tool surface covers key operations: status, listing resources, fetching data, refreshing public data, and executing actions. Minor potential gaps like direct search or filtering are not critical given the design.

    Maintenance

    ActivityNo data
    ResponsivenessOutdated data

    Related MCP Connectors

    Related MCP Servers

    • A
      license
      Not graded
      quality
      D
      maintenance
      MCP server that provides AI agents with financial tools including real-time quotes, backtesting, technical analysis, and multi-exchange data via a simple CLI interface.
      2
      MIT
    • A
      license
      Not graded
      quality
      D
      maintenance
      Enables AI agents to interact with Interactive Brokers through 48 tools for market data, orders, account management, and more, via the MCP protocol.
      MIT