by AllPath
AllPath · Trading Agent
Self-hosted, open source. Write strategies in plain YAML, let a sentinel watch them every hour, chat with an agent that remembers how you think — and keep a human approval gate in front of every single order.
How an order happens
rule fires
price < 150 in your YAML strategy
agent researches
reads the chart, your memory, the thesis
proposal queued
order or strategy change, with a diff
you approve
web, phone push, or Telegram link
risk gate
deterministic caps — cannot be bypassed
paper order
Alpaca paper account by default
The agent has no tool that places an order or writes a strategy file. It can research, remember, and propose. Approval is a click on the Pending page, a one-time link in a notification, or nothing at all — in which case nothing happens.
Strategy sentinel
Strategies are plain YAML: a thesis, a target weight, and rules like `price < 140 → sell all`. During market hours the sentinel evaluates every active strategy on your interval, arms and fires rules, and queues anything that needs a decision.
An agent that remembers
Ask what a position looks like against its plan, tighten a stop, draft a new strategy. It keeps four layers of memory (your profile, strategies, per-stock notes, lessons) and consolidates them nightly, so it gets less generic the longer you use it. Bring your own model: OpenRouter (any model, one key), OpenAI, or Anthropic — pick a different tier for chat, hourly review, and memory.
Nightly reflection
A bounded reflection session reviews every strategy against the day's fills, prices, and triggers, writes durable lessons into memory, and — when a thesis has drifted — proposes a revision. You get the report on your phone; the proposal waits on the Pending page.
What it will not do
Paper by default
Alpaca paper trading out of the box. Live trading is a deliberate .env change, not a checkbox.
Every order waits for you
The agent can only propose. Web, one-tap link, or Telegram — but a human clicks Approve.
Risk gate in code
Position caps and daily limits are deterministic Python the LLM cannot reach around.
Strategy edits are diffs
Chat drafts and reflection proposals land as side-by-side diffs. Nothing writes to a strategy file except your approval.
Runs on your machine
SQLite, local YAML, your own API keys. Nothing leaves your box but the calls you configured.
Honest state
Draft strategies are labelled not monitored. Fills show submitted vs filled. A stale heartbeat shows as stale.
Quick start
Or: one prompt, zero steps
Using Claude Code, Codex, Cursor — any coding agent? Paste this and it does the setup for you.
Set up the AllPath Trading Agent (https://github.com/dukesky/allpath-trading-agent) on this machine and hand me a running dashboard: 1. Check Python 3.11+ is available and that uv is installed (install uv if missing: https://docs.astral.sh/uv/getting-started/installation/). 2. git clone https://github.com/dukesky/allpath-trading-agent, cd into it, run: uv sync 3. cp .env.example .env — leave every key EMPTY. Do not ask me for API keys or secrets in this chat, and never commit .env. 4. Start the server as a background process: uv run allpath-trade serve 5. First run prints a one-time login token in the server output — find it and show it to me. 6. Open http://127.0.0.1:8791 in my browser. I'll sign in with the token; the built-in setup wizard then collects my Alpaca paper keys (free account, no funding needed) and the Settings page takes one LLM key (OpenRouter, OpenAI, or Anthropic). Everything runs locally on 127.0.0.1 and it is paper trading by default — do not change that. You are done when the dashboard loads and you've shown me the login token.
Your API keys never touch the agent chat — the app's own setup wizard collects them locally afterwards. Prefer doing it by hand? The same six steps are below.
Clone and install
Needs Python 3.11+ and uv. Nothing else to install — the broker and data clients ship with the package.
git clone https://github.com/dukesky/allpath-trading-agent cd allpath-trading-agent uv sync
Get a broker key (free paper account)
Alpaca paper accounts are free and need no funding. Generate the key pair from the Paper Trading dashboard (right-hand panel).
# .env ALPACA_API_KEY=your_paper_key ALPACA_SECRET_KEY=your_paper_secret ALPACA_PAPER=true
Pick an LLM provider
One key is enough. OpenRouter is recommended (any model behind one key); OpenAI and Anthropic connect directly. You can pick different models per tier later in Settings.
# .env — choose ONE LLM_PROVIDER=openrouter OPENROUTER_API_KEY=sk-or-... # or: LLM_PROVIDER=openai + OPENAI_API_KEY=... # or: LLM_PROVIDER=anthropic + ANTHROPIC_API_KEY=...
Start the web UI
The first run prints a login token in the terminal. Open the URL, paste the token, and you are on the dashboard.
cp .env.example .env # then fill in the keys above uv run allpath-trade serve # → http://127.0.0.1:8791
Draft a strategy in chat, then approve it
Describe the plan in plain English. The draft lands on Pending as a diff; approve it, then activate it on the Strategies page so the sentinel starts watching.
You: "Buy NVDA on pullbacks under 170, up to 20% of the account, hard stop at 140." Agent: Draft queued for your approval as #1 — open Pending. # Approve on Pending → Activate on the Strategies page → the sentinel starts watching
Take it with you (optional)
Push via ntfy or pair a Telegram bot in Settings. Triggers, fills, nightly reports, and one-tap approve links reach your phone.
Settings → Push notifications (ntfy) or Settings → Telegram # triggers, fills, nightly reports and approve links on your phone
Full setup, the strategy YAML reference, and every safety note are in the README (English and Chinese). Live trading is a deliberate .env change, never the default.
More of the UI
