Stock Tracker — Market Context Brief & Anomaly Alerts

Automated stock monitoring system built with n8n + Claude Haiku + Warren (OpenClaw agent) + a Python anomaly-detection layer (market_intelligence/).

Three independent layers, one Telegram channel:

Layer	What	When	LLM?
Macro Brief	Daily market context brief — Fed, rates, dollar, oil, VIX, small caps, sectors, geopolitics	16:00 Paris, Mon–Fri	Haiku (web search) + Warren (prose)
A — News (silent)	Web news per ticker → dedup → memory write only, no message	16:00 Paris, Mon–Fri	Haiku (search) + Warren (filter)
B — Anomaly detection (EOD)	Price/volume anomaly scan → beta gate → dedup → targeted Warren explanation	21:30 UTC, Mon–Fri (after US close)	Zero LLM in the detection path — Warren only called for surviving alerts

The Macro Brief answers "what is the market doing today?". Layer A silently maintains the news memory that feeds Layer B context. Layer B answers "when" (something unusual just happened on this stock).

Macro Brief — Daily Market Context

Macro Brief Schedule 16h Mon-Fri (Europe/Paris)
         │
         ▼
  Call Warren Macro Brief — POST /macro-brief
  (warren_server.py — handle_macro_brief)
         │
         ├─ FRED API (get_snapshot)
         │   Fed Funds, 10Y/2Y rates, VIX, dollar index, S&P 500
         │
         ├─ yfinance (get_market_closes)
         │   IWM close + daily change, WTI crude close + daily change
         │
         ├─ Haiku web_search (fetch_macro_snapshot)
         │   Fed stance, rate expectations/rumors, geopolitics,
         │   notable IPOs, hot sectors, Fear & Greed, market rumors
         │
         ├─ sector_rotation.py (get_sector_rotation) — zero LLM
         │   Sector ETF relative performance, IWM/SPY ratio,
         │   small caps appetite trend
         │
         └─ fear_greed.py (get_fear_greed) — zero LLM
             CNN Fear & Greed index (best-effort)
         │
         ▼
  Warren — prose brief 150–300 words
  Signal-first: opens with dominant regime (risk-on / risk-off / neutral)
  Max 5 numeric values — Fed, rates, VIX, IWM move, Fear & Greed
  No bullet points, no tables, no # headings, no bold headers
  Rumors always labeled as rumors
  Closes with one-sentence regime conclusion
         │
         ▼
  Extract Macro Brief
         │
         ▼
  Aggregate for Telegram → Split for Telegram → Send Telegram

Graceful degradation — if one source fails, the brief continues with the remaining data:

Web search fails → quantitative FRED data only, brief mentions qualitative uncertainty
All FRED sources fail → fallback values + warning
Brief is never empty, never crashes

Estimated cost: ~$0.02/day (Haiku macro web search + Warren brief call)

Layer A — Silent News Collection

Layer A News Schedule 16h Mon-Fri (Europe/Paris)
         │
         ▼
  Read Tickers — reads portfolio.json + watchlist.json (16 tickers: 8 + 8)
         │
         ▼  × 16 parallel calls
  Claude Haiku + web_search
  "Any news on TICKER from TODAY?"
  max 3 searches · 512 tokens
         │
         ▼
  Aggregate all raw news
         │
         ▼
  Call Warren Filter — ticker-watch  (POST /filter)
  Reads memory/tickers/SYMBOL.md (last 3 entries)
  Returns { new: [...], skip: [...] }
         │
         ▼
  Extract Filter Result → If New Items
         │
         ▼ NEW tickers only
  Call Memorize — POST /memorize (handle_memorize)
  Writes memory/tickers/SYMBOL.md for each NEW ticker
  No Warren synthesis call — no Telegram message

SKIP logic — a ticker is skipped if:

Haiku found no news from today
Today's news is semantically identical to an existing memory entry (duplicate)

If no ticker is NEW → pipeline stops silently. No message is sent.

Purpose — Layer A feeds memory/tickers/SYMBOL.md, which Warren reads when researching anomaly alerts (Layer B step S7). Suppressing the synthesis/Telegram node removes noise while preserving the memory pipeline.

Note on /synthesize — warren_server.py still exposes POST /synthesize (legacy endpoint, present for backward compatibility). It is no longer called by the n8n workflow.

Layer B — EOD anomaly pipeline

Layer B EOD Schedule (21:30 UTC, Mon–Fri — DST-safe, always ≥ 30 min after US close)
         │
         ▼
  python3 -m market_intelligence.eod_orchestrator --history-days 280
         │   S0 fetch EOD OHLCV (yfinance, Twelve Data fallback) + registry/quarantine
         │   S1 anomaly signals (z-score MAD, RVOL, gap, ATR, 52-week breakout)
         │   S2 beta gate (market-model regression vs IWM + sector ETF)
         │   S3 candidate alerts (calm |z|>2.0 / speculative |z|>2.5)
         │   S4 short interest → squeeze-prone flag
         │   S5 dedup (hysteresis latch per ticker)
         │   S6 macro snapshot (computed once, cached, attached to every alert)
         │   S7 Warren targeted research per surviving alert:
         │      EDGAR Form 4 (structured) · product/sector news (yfinance)
         │      halt status (FINRA) · SSR status (Nasdaq) · squeeze flag
         │      Layer A news memory for the ticker · macro snapshot
         │      → explicitly allowed to answer "no identifiable catalyst"
         ▼
  JSON result { survivor_count, should_send, digest, data_issues }
         │
         ▼ if survivors
  One Telegram digest (reuses Aggregate for Telegram / Split for Telegram / Send Telegram nodes)

No alert survives → nothing is sent.

🔍 Anomaly detection to anticipate a potential move

The problem we are solving

On speculative small caps, price often moves before the news becomes public (rumor, accumulation, squeeze). A pipeline that only reads news therefore arrives late. Layer B does not predict anything: it detects that an unusual move has just happened, then asks Warren to investigate. It is an attention detector, not a crystal ball.

Step 1 — "Is this move unusual for THIS ticker?" (the z-score)

A stock moving 8% in one day is huge for Xylem (a quiet water company) and ordinary for Rigetti (a volatile quantum stock). A fixed percentage threshold would therefore make no sense. Instead, the day's move is compared with the ticker's own behavior over the last 60 days:

z-score = "how many times larger than usual?" z = 1 → normal day. z = 2 → big, rare day. z = 3 → exceptional.

The z-score "normalizes" each ticker by its own volatility.

Robustness detail: "usual behavior" is measured with the median (MAD) rather than the mean, so that a few past extreme sessions do not distort the reference.

Step 2 — "Is the ticker moving, or is the whole market moving?" (the beta gate)

If the whole small-cap market drops 3%, your speculative ticker may drop 5% without any company-specific news. Alerting on that would be noise. The image: the tide vs. the wave. We want to detect the wave (the ticker's own move), not the tide (the whole market's move).

In practice, the system learns the ticker's market sensitivity over 60 days (its "beta": when the IWM small-cap index moves 1%, this ticker moves 2% on average). On the day of the move, it calculates the expected move given what the market did, subtracts it from the real move, and keeps the residual: the part of the move the market does not explain. The z-score from step 1 is applied to this residual.

For strongly thematic stocks (nuclear, quantum, water...), the system also removes the part explained by the sector ETF (NUKZ for SMR/OKLO, QTUM for RGTI...), but only if the ticker actually follows that ETF (correlation > 0.35); otherwise it would only introduce noise.

A price alert = |residual z| > 2 (calmer names: XYL, MMED) or > 2.5 (speculative cluster: RGTI, BBAI, OKLO...). One global setting; per-ticker normalization does the rest.

Step 3 — The other sensors

RVOL (relative volume): today's volume ÷ 20-day average volume. Volume ×3 with no news = someone may know something, or a squeeze may be starting. This is the #1 early signal on small caps.
ATR expansion: the day's range (high-low) exceeds 1.5× the usual range → "the ticker is waking up", even if it closes flat.
52-week breakout: new yearly high or low.
Combination: alert if price is abnormal, OR if abnormal volume + a second signal confirms it.

Step 4 — Avoid crying wolf twice (hysteresis)

A stock that takes off often remains volatile for several days. Without a guardrail, the same alert would arrive every evening. The mechanism works like a thermostat: once the alert fires, the ticker is "locked" and does not re-alert until it has calmed down again (|z| < 1 for at least one day), unless something genuinely new happens: direction reversal, a new signal type, or a clear escalation. Safety valve: after about 10 trading days, the lock expires.

Step 5 — Warren investigates (and is allowed to find nothing)

Only surviving alerts cost an LLM call. Warren receives a structured dossier: insider buying/selling (SEC EDGAR Form 4), product and sector news, trading suspension status (FINRA), short-sale restriction status (Nasdaq), a "squeeze-prone ticker" flag (short interest), Layer A's news memory for that ticker, and the day's macro context. Warren must explain the move without making things up: the prompt explicitly allows it to conclude "no identifiable catalyst — flow/technical/squeeze likely".

What anomaly detection does not do

It does not predict direction (a volume spike says "look", not "it goes up").
It does not trade. A high false-positive rate is accepted: this is an attention tool, strictly better than a daily news scan, not a robot.

Configuration: thresholds in market_intelligence/data/alert_thresholds.json, sector mapping in data/sector_factors.json, hysteresis in data/dedup_thresholds.json, squeeze in data/short_interest_thresholds.json.

Skills — on-demand Telegram commands

Skill	Trigger	What
tickerbrief	`brief TICKER`, `point sur TICKER`, `actu TICKER`	On-demand brief: today's news + ticker memory + EOD anomaly state + sector. Read-only.
modifyportfolio	natural language via Warren	Add/remove tickers from portfolio.json interactively
modifywatchlist	natural language via Warren	Add/remove tickers from watchlist.json interactively

tickerbrief assembles: memory/tickers/SYMBOL.md, fresh web search news, dedup_state.json anomaly state, sector_factors.json. Read-only — never writes files. Returns a signal-first Telegram reply (anomaly status first, then news, then memory). For tickers not in portfolio/watchlist, returns raw web search only.

Features

Macro Brief — daily market regime brief sent at 16h Paris even with no ticker news
Two independent layers — silent news collection (A) + EOD anomaly trigger (B), one Telegram channel
Zero LLM in the detection path — Layer B is pure Python/statistics until an alert survives
Date-filtered news — Haiku searches for today's news only
Duplicate memory — Warren compares each ticker's news against the last 3 entries; only new content is written
Layer A ↔ B cross-reference — Warren's anomaly research includes the ticker's news memory
Sector rotation signal — sector_rotation.py computes sector ETF relative performance + IWM/SPY ratio (zero LLM)
Fear & Greed — fear_greed.py fetches CNN Fear & Greed index (best-effort, zero LLM)
On-demand ticker brief — tickerbrief skill via Telegram: full ticker context without triggering a digest
Signal-first briefing — dominant market regime on the first line of the macro brief
Tickers as data — portfolio.json / watchlist.json are the source of truth, editable via Telegram (modifyportfolio / modifywatchlist skills)
Symbol integrity — registry + quarantine (market_intelligence/data/) so analysis never runs on a wrong ticker
Auto-split Telegram — messages split at 4,000 characters at paragraph boundaries
systemd managed — n8n, OpenClaw gateway, Warren HTTP bridge
CI/CD — push to main → tests → auto-deploy on the VPS via self-hosted runner

Stack

Component	Role
n8n (self-hosted)	Scheduling, API calls, credential management, delivery
Claude Haiku (`claude-haiku-4-5-20251001`)	Per-ticker raw news search via `web_search` + macro web search
OpenClaw	Agent framework wrapping Claude for Warren
Warren (OpenClaw agent)	Intelligence layer — filtering, memory, Macro Brief prose, alert explanations
warren_server.py	Python HTTP bridge between n8n and OpenClaw CLI (port 18795)
agents/warren/	Prompt builder (persona, output format) + macro providers (FRED, web search, market closes)
market_intelligence/	Layer B — EOD fetch, anomaly signals, beta gate, dedup, EDGAR, short interest, orchestrator
market_intelligence/sector_rotation.py	Sector ETF relative performance + IWM/SPY ratio — zero LLM, feeds Macro Brief
market_intelligence/fear_greed.py	CNN Fear & Greed index fetch — zero LLM, feeds Macro Brief

Project structure

stock-tracker/
├── workflow.json              # n8n workflow (Macro Brief + Layer A silent + Layer B wiring)
├── warren_server.py           # Python HTTP bridge (port 18795): /filter /memorize /macro-brief /synthesize(legacy)
├── portfolio.json             # 8 portfolio tickers (source of truth)
├── watchlist.json             # 8 watchlist tickers (source of truth)
├── requirements.txt           # Python deps (pydantic, requests, anthropic, numpy, pandas, yfinance, pyarrow)
├── agents/warren/             # Prompt builder, macro providers (FRED, web search, market closes), ticker management
├── market_intelligence/       # Layer B anomaly detection (S0–S8)
│   ├── sector_rotation.py     # Sector ETF rotation + IWM/SPY ratio (zero LLM)
│   ├── fear_greed.py          # CNN Fear & Greed index (zero LLM)
│   └── data/                  # registry, quarantine, thresholds, sector factors
├── skills/                    # OpenClaw skills sources
│   ├── macrobrief/            # Macro Brief skill spec (SKILL.md)
│   ├── tickerbrief/           # On-demand ticker brief skill spec (SKILL.md)
│   ├── modifyportfolio/       # Portfolio management via Telegram
│   └── modifywatchlist/       # Watchlist management via Telegram
├── tests/                     # pytest suite (agents, market_intelligence, workflow wiring)
├── docs/                      # project-structure, deployment, ticker schema
├── deploy/                    # CI/CD: remote.sh + import_workflow.py
└── .github/workflows/         # CI + auto-deploy + Notion sync

/home/warren/.openclaw/workspace-warren/      (on the VPS)
├── PROMPT.md / SOUL.md / IDENTITY.md / ...   # Warren agent definition
├── ARCHITECTURE.md                            # Pipeline documentation
├── skills/
│   ├── ticker-watch/          # Filter skill (NEW vs SKIP)
│   ├── macro-brief/           # Macro Brief skill (French prose, 150-300 words)
│   ├── tickerbrief/           # On-demand ticker brief skill
│   ├── modifyportfolio/       # Telegram portfolio management
│   └── modifywatchlist/       # Telegram watchlist management
└── memory/tickers/            # SYMBOL.md — last 3 raw news entries (written by Layer A /memorize)

VPS Installation (Ubuntu 22.04+)

Prerequisites

Ubuntu 22.04 VPS (2 GB RAM minimum)
A dedicated system user for the app (recommended: warren)
Your Anthropic API key, Telegram bot token

1. Create system user

sudo useradd -m -s /bin/bash warren
sudo usermod -aG sudo warren  # optional — remove after setup if desired

2. Install Node.js

# As warren user (or any user that will run n8n)
curl -o- https://raw.githubusercontent.com/nvm-sh/nvm/v0.39.7/install.sh | bash
source ~/.bashrc
nvm install 22
node -v   # v22.x.x

Or system-wide via NodeSource:

curl -fsSL https://deb.nodesource.com/setup_22.x | sudo -E bash -
sudo apt-get install -y nodejs

3. Install n8n

sudo npm install -g n8n
n8n --version

4. Install OpenClaw

OpenClaw is the agent framework that runs Warren. Install it globally:

sudo npm install -g openclaw
# or: npm install -g @openclaw/cli  (check current package name)
openclaw --version

Then configure OpenClaw for the warren user:

sudo -u warren openclaw init

This creates /home/warren/.openclaw/ with the default config.

5. Clone the repository and install Python deps

sudo mkdir -p /opt/apps/stock-tracker
sudo chown warren:warren /opt/apps/stock-tracker
sudo -u warren git clone https://github.com/patw47/stock-tracker.git /opt/apps/stock-tracker
cd /opt/apps/stock-tracker
sudo pip3 install --break-system-packages -r requirements.txt

anthropic is required by the macro web search (Haiku web_search calls for Fed/geopolitics/sectors). Without it the Macro Brief silently falls back to quantitative FRED data only.

6. Configure environment variables

sudo -u warren cp .env.example .env
sudo -u warren nano .env

Fill in:

ANTHROPIC_API_KEY=sk-ant-...
TELEGRAM_TOKEN=123456789:ABC...
TELEGRAM_CHAT_ID=1234567890
TWELVE_DATA_API_KEY=...               # Layer B EOD fallback data source
NODE_FUNCTION_ALLOW_BUILTIN=fs        # lets n8n Code nodes read portfolio/watchlist.json
N8N_PORT=5680
N8N_USER_FOLDER=/opt/apps/stock-tracker/n8n-data
N8N_BASIC_AUTH_ACTIVE=true
N8N_BASIC_AUTH_USER=admin
N8N_BASIC_AUTH_PASSWORD=your-password
GENERIC_TIMEZONE=Europe/Paris
N8N_DEFAULT_LOCALE=fr
N8N_ENCRYPTION_KEY=$(openssl rand -base64 32)

Telegram: create a bot via @BotFather, get your chat ID via @userinfobot.

7. Set up Warren workspace

Copy the workspace files to the OpenClaw directory:

sudo cp -r /opt/apps/stock-tracker/workspace-warren \
  /home/warren/.openclaw/workspace-warren
sudo chown -R warren:warren /home/warren/.openclaw/workspace-warren

Create the memory directory:

sudo -u warren mkdir -p /home/warren/.openclaw/workspace-warren/memory/tickers

Register Warren as an OpenClaw agent:

sudo -u warren openclaw agent add warren \
  --workspace /home/warren/.openclaw/workspace-warren \
  --model claude-haiku-4-5-20251001

Verify:

sudo -u warren openclaw agent list   # warren should appear

8. Configure OpenClaw with your Anthropic key

sudo -u warren openclaw config set anthropic.apiKey sk-ant-...
# or edit /home/warren/.openclaw/openclaw.json directly

9. Set up systemd services

n8n service

sudo nano /etc/systemd/system/stock-tracker.service

[Unit]
Description=Stock Tracker — n8n
After=network.target

[Service]
Type=simple
User=warren
WorkingDirectory=/opt/apps/stock-tracker
EnvironmentFile=/opt/apps/stock-tracker/.env
ExecStart=/usr/bin/n8n start
Restart=always
RestartSec=10

[Install]
WantedBy=multi-user.target

OpenClaw gateway service

sudo nano /etc/systemd/system/openclaw-warren.service

[Unit]
Description=OpenClaw Warren Gateway
After=network.target

[Service]
Type=simple
User=warren
Environment=HOME=/home/warren
Environment=PATH=/usr/local/bin:/usr/bin:/bin
ExecStart=/usr/local/bin/openclaw gateway start --agent warren
Restart=always
RestartSec=5

[Install]
WantedBy=multi-user.target

Warren HTTP bridge service

sudo nano /etc/systemd/system/warren-server.service

[Unit]
Description=Warren HTTP Bridge — n8n to OpenClaw
After=network.target openclaw-warren.service
Wants=openclaw-warren.service

[Service]
Type=simple
User=warren
Environment=HOME=/home/warren
Environment=PATH=/usr/local/bin:/usr/bin:/bin
EnvironmentFile=/opt/apps/stock-tracker/.env
ExecStart=/usr/bin/python3 /opt/apps/stock-tracker/warren_server.py
Restart=always
RestartSec=5

[Install]
WantedBy=multi-user.target

EnvironmentFile is required on the bridge service: the Macro Brief calls the Anthropic API directly (Haiku web search) and needs ANTHROPIC_API_KEY in the process environment.

Enable and start all three:

sudo systemctl daemon-reload
sudo systemctl enable stock-tracker openclaw-warren warren-server
sudo systemctl start openclaw-warren warren-server
sleep 3
sudo systemctl start stock-tracker

Verify:

sudo systemctl status stock-tracker openclaw-warren warren-server

All three should show active (running).

10. Import the workflow into n8n

sudo -u warren \
  N8N_USER_FOLDER=/opt/apps/stock-tracker/n8n-data \
  N8N_ENFORCE_SETTINGS_FILE_PERMISSIONS=false \
  n8n import:workflow --input=/opt/apps/stock-tracker/workflow.json

Activate the workflow via SQLite (n8n import deactivates by default):

sqlite3 /opt/apps/stock-tracker/n8n-data/.n8n/database.sqlite \
  "UPDATE workflow_entity SET active=1 WHERE id='veille-boursiere-001';"

Restart n8n to pick up the change:

sudo systemctl restart stock-tracker

11. Configure credentials in the n8n UI

Access the UI via SSH tunnel (n8n binds to localhost only):

ssh -L 5680:localhost:5680 warren@your-vps-ip -N

Then open http://localhost:5680 in your browser.
Login: the values you set in .env (N8N_BASIC_AUTH_USER / N8N_BASIC_AUTH_PASSWORD).

Go to Credentials and create:

Name	Type	Settings
Header Auth account	HTTP Header Auth	Header: `x-api-key` · Value: your Anthropic key
Telegram account	Telegram API	Token from BotFather

After creating credentials, check the node IDs in the workflow match. If they don't, open each affected node and reselect the credential from the dropdown.

12. Smoke test

Test the Warren HTTP bridge directly:

# Test filter endpoint
curl -s -X POST http://127.0.0.1:18795/filter \
  -H 'Content-Type: application/json' \
  -d '{"news":{"OKLO":"NO_NEWS_TODAY","SMR":"NuScale signs PPA with Azure today"}}' \
  | python3 -m json.tool

# Expected: {"new": ["SMR"], "skip": ["OKLO"], "reasons": {...}}

# Test memorize endpoint (silent memory write, no synthesis)
curl -s -X POST http://127.0.0.1:18795/memorize \
  -H 'Content-Type: application/json' \
  -d '{"newTickers":["SMR"],"allNews":{"SMR":"NuScale signs PPA with Azure today"}}' \
  | python3 -m json.tool

# Expected: {"status": "ok", "written": ["SMR"]}

# Test macro-brief endpoint
curl -s -X POST http://127.0.0.1:18795/macro-brief \
  -H 'Content-Type: application/json' \
  -d '{}' \
  | python3 -m json.tool

# Expected: {"brief": "Le marché aborde cette séance dans un régime..."}

Test the Layer B pipeline end-to-end (prints a JSON payload):

cd /opt/apps/stock-tracker && python3 -m market_intelligence.eod_orchestrator --history-days 280

Test n8n is up:

curl -s http://localhost:5680/healthz
# Expected: {"status":"ok"}

Continuous deployment (GitHub Actions → VPS)

Pushing to main auto-deploys to the VPS once CI is green. The deploy job runs on a self-hosted GitHub Actions runner installed on the VPS (no SSH).

push main → CI green ──► .github/workflows/deploy.yml  (runs-on: self-hosted)
                              │
                              ├─ gate: "Notify PR merge conflicts" green on same commit
                              ├─ git fetch (token-auth) + reset --hard origin/main
                              └─ deploy/remote.sh (local on the VPS):
                                   pip install -r requirements.txt
                                   stop stock-tracker  (release sqlite lock + free port 5679)
                                   import_workflow.py  (upsert workflow into sqlite)
                                   restart openclaw-warren → warren-server
                                   n8n execute --id  (validation run, as warren, n8n stopped)
                                   start stock-tracker
                                   healthcheck: services + n8n :5680/healthz + bridge :18795
                              → Telegram status message (success / failure + per-service state)

See docs/deployement.md for the full rationale (self-hosted runner, custom sqlite importer, validation-run constraints, user roles).

Customizing tickers

portfolio.json and watchlist.json are the source of truth — the n8n Read Tickers node and the Layer B registry read them at runtime. Three ways to edit:

Telegram — talk to Warren: the modifyportfolio / modifywatchlist skills add/remove tickers interactively (inline buttons, confirmation message).
Edit the JSON files on the VPS (/opt/apps/stock-tracker/*.json) — picked up at the next run, no n8n change needed.
Git — commit the change; deploy syncs the files.

Each entry needs symbol, name, sector (see docs/ticker-files-schema.md). New tickers are validated against the Layer B registry; unresolvable symbols are quarantined (market_intelligence/data/quarantine.json) instead of corrupting analysis.

Warren memory

Warren stores raw news per ticker in /home/warren/.openclaw/workspace-warren/memory/tickers/.

One file per ticker: SYMBOL.md
Max 3 entries, newest first, separated by ---
Written by Layer A /memorize endpoint — only when new content is confirmed by Warren filter
Also read by Layer B: the anomaly research prompt includes this memory, so Warren interprets a price move knowing what news already surfaced for the ticker
Also read by tickerbrief skill: on-demand context without triggering a new search
To reset a ticker's memory: rm memory/tickers/SYMBOL.md
To reset all memory: rm memory/tickers/*.md

Service management

# Status
sudo systemctl status stock-tracker warren-server openclaw-warren

# Restart all
sudo systemctl restart openclaw-warren warren-server stock-tracker

# Logs
sudo journalctl -u warren-server -f       # Python bridge logs
sudo journalctl -u stock-tracker -f       # n8n logs
sudo journalctl -u openclaw-warren -f     # OpenClaw logs

Environment variables reference

Variable	Description
`ANTHROPIC_API_KEY`	Anthropic API key (Haiku web_search in n8n + macro brief Haiku calls in the bridge)
`TELEGRAM_TOKEN`	Telegram bot token
`TELEGRAM_CHAT_ID`	Target chat ID
`TWELVE_DATA_API_KEY`	Layer B EOD data fallback (yfinance primary)
`NODE_FUNCTION_ALLOW_BUILTIN`	Must include `fs` — n8n Code nodes read the ticker JSON files
`N8N_PORT`	n8n HTTP port (default: `5680`)
`N8N_USER_FOLDER`	n8n data directory
`N8N_BASIC_AUTH_USER`	n8n login username
`N8N_BASIC_AUTH_PASSWORD`	n8n login password
`N8N_ENCRYPTION_KEY`	Credentials encryption key (generate once, never change)
`GENERIC_TIMEZONE`	Timezone for Layer A and Macro Brief scheduling (`Europe/Paris`). Layer B cron is UTC by design (DST safety)
`GMAIL_USER` / `GMAIL_PASS`	Legacy — email delivery was removed; Telegram only

Architecture decisions

Specs and decision log live in the Notion epics database (« Epics Stock Tracker ») and the Obsidian vault (Memory/stock-tracker/epics/). Key choices:

Macro Brief as main product — daily market regime brief even with no ticker news; Layer A becomes a silent collector
Layer A silent collection — the Haiku + filter + memory pipeline survives suppression of synthesis/Telegram; preserves the ticker_news_memory context consumed by Warren S7
/memorize endpoint decoupled from /synthesize — memory write triggered by Layer A without a Warren synthesis call; /synthesize kept as legacy for backward compatibility
sector_rotation.py + fear_greed.py zero-LLM — quantitative signals injected into the Macro Brief without adding LLM cost to the collection path
Haiku for search — cheaper, faster, sufficient for raw news retrieval and macro web search
Warren filter only in Layer A — filter (NEW/SKIP dedup) retained; synthesis removed
Memory = raw news — storing Haiku output (not Warren synthesis) for stable duplicate comparison
warren_server.py — Python bridge needed because n8n sandboxes fs and child_process modules
No LLM in the detection path — Layer B anomalies are pure statistics; Warren is only paid for surviving alerts
Beta gate = market-model regression (not naive z-score comparison) — avoids false alerts on broad risk-off days for high-beta names
MAD scale, not standard deviation — robust to the fat tails of speculative small-caps
Hysteresis dedup — one alert per event, not per day; re-arms when the ticker calms down
Layer B cron in fixed UTC (21:30) — Paris-time cron ran before the US close for ~3 weeks each March (EU/US DST mismatch)
systemd over PM2 — PM2 not available on this VPS; systemd provides equivalent reliability

Name		Name	Last commit message	Last commit date
Latest commit History 140 Commits
.claude/agents		.claude/agents
.codex		.codex
.github/workflows		.github/workflows
agents		agents
deploy		deploy
docs		docs
market_intelligence		market_intelligence
skills		skills
tests		tests
.conventions		.conventions
.env.example		.env.example
.epic.md		.epic.md
.gitignore		.gitignore
.rules		.rules
.sprint.md		.sprint.md
AGENTS.md		AGENTS.md
CLAUDE.md		CLAUDE.md
README.md		README.md
call_warren.sh		call_warren.sh
ecosystem.config.js		ecosystem.config.js
portfolio.example.json		portfolio.example.json
requirements.txt		requirements.txt
start.sh		start.sh
warren_server.py		warren_server.py
watchlist.example.json		watchlist.example.json
workflow.json		workflow.json

Folders and files

Latest commit

History

Repository files navigation

Stock Tracker — Market Context Brief & Anomaly Alerts

Macro Brief — Daily Market Context

Layer A — Silent News Collection

Layer B — EOD anomaly pipeline

🔍 Anomaly detection to anticipate a potential move

The problem we are solving

Step 1 — "Is this move unusual for THIS ticker?" (the z-score)

Step 2 — "Is the ticker moving, or is the whole market moving?" (the beta gate)

Step 3 — The other sensors

Step 4 — Avoid crying wolf twice (hysteresis)

Step 5 — Warren investigates (and is allowed to find nothing)

What anomaly detection does not do

Skills — on-demand Telegram commands

Features

Stack

Project structure

VPS Installation (Ubuntu 22.04+)

Prerequisites

1. Create system user

2. Install Node.js

3. Install n8n

4. Install OpenClaw

5. Clone the repository and install Python deps

6. Configure environment variables

7. Set up Warren workspace

8. Configure OpenClaw with your Anthropic key

9. Set up systemd services

n8n service

OpenClaw gateway service

Warren HTTP bridge service

10. Import the workflow into n8n

11. Configure credentials in the n8n UI

12. Smoke test

Continuous deployment (GitHub Actions → VPS)

Customizing tickers

Warren memory

Service management

Environment variables reference

Architecture decisions

About

Resources

Uh oh!

Stars

Watchers

Forks

Releases

Packages 0

Uh oh!

Contributors

Uh oh!

Languages

Packages