Kraken Grid Trader

For Claude: How to Use This Skill

Skill instructions are preloaded in context when this skill is active. Do not perform filesystem searches or tool-driven exploration to rediscover them; use the guidance below directly.

Windows / Seren Desktop Python Runtime

Seren Desktop bundles Python on Windows and prepends the bundled runtime to child-process PATH. When running from Seren Desktop on Windows, use the documented python3 ... commands as written; they resolve to Seren Desktop's bundled python3.exe even when system Python is not installed. Do not translate python3 to python, do not invoke the Microsoft Store Python stub, and do not ask the user to install system Python just to run this skill from Seren Desktop. If a snippet includes Unix virtualenv activation such as source .venv/bin/activate, skip that prefix inside Windows Desktop and run the same command beginning with python3 .... Outside Seren Desktop, use an installed Python 3.11+ interpreter or a project .venv.

Automated grid trading bot for Kraken that profits from BTC volatility using a mechanical, non-directional strategy.

On Invoke

Immediately run a dry-run grid simulation without asking. Do not present a menu of modes. Execute:

cd ~/.config/seren/skills/grid-trader && source .venv/bin/activate && python3 scripts/agent.py dry-run --config config.json --cycles 5

Display the full dry-run results to the user. Only after results are displayed, present available next steps (live mode with --allow-live). If the user explicitly requests a specific mode in their invocation message, run that mode instead.

What This Skill Provides

• Automated Kraken grid trading with dry-run and live modes
• Pair selection support (single pair or candidate list)
• Adaptive grid centering, spacing/order-size tuning, and persistent learning state
• Shadow evaluation with gated promotion and rollback
• JSONL logs for setup, orders, fills, positions, and errors
• MCP-native SerenDB persistence for sessions, events, orders, fills, position snapshots, adaptive runtime state, telemetry, reviews, and runtime locks

What is Grid Trading?

Grid trading places buy and sell orders at regular price intervals (the "grid"). When price moves up and down, orders fill automatically — accumulating profit from oscillation without predicting direction.

API Key Setup

Before running this skill, check for an existing Seren API key in this order:

1. Seren Desktop auth — if the skill is running inside Seren Desktop, the runtime injects API_KEY automatically. Check: echo $API_KEY. If set, no further action is needed.
2. Existing .env file — check if SEREN_API_KEY is already set in the skill's .env file. If set, no further action is needed.
3. Shell environment — check if SEREN_API_KEY is exported in the current shell. If set, no further action is needed.

Only if none of the above are set, register a new agent account:

curl -sS -X POST "https://api.serendb.com/auth/agent" \
  -H "Content-Type: application/json" \
  -d '{"name":"grid-trader"}'

Extract the API key from the response at .data.agent.api_key — this key is shown only once. Write it to the skill's .env file:

SEREN_API_KEY=<the-returned-key>

Verify:

curl -sS "https://api.serendb.com/auth/me" \
  -H "Authorization: Bearer $SEREN_API_KEY"

Do not create a new account if a key already exists. Creating a duplicate account results in a $0-balance key that overrides the user's funded account.

Reference: https://docs.serendb.com/skills.md

Setup

1. Configure Kraken publisher credentials in Seren Desktop Settings → Publisher MCPs (desktop sidecar/keychain flow)
2. Copy .env.example to .env and set SEREN_API_KEY
3. Copy config.example.json to config.json and configure your grid parameters
4. Install dependencies: pip install -r requirements.txt
5. Run: python scripts/agent.py

Trade Execution Contract

When the user says sell, close, exit, unwind, or flatten, treat that as an immediate operator instruction to stop new grid entries and cancel open Kraken orders for the configured pair. If the user did not identify which pair or campaign to stop, ask only the minimum clarifying question needed to identify it.

Pre-Trade Checklist

Before any live start --allow-live run:

1. Fetch balances and the latest Kraken price for the active pair.
2. Verify SEREN_API_KEY and Kraken publisher credentials are loaded.
3. Verify grid spacing, quote reserve, position size, and drawdown caps still fit the account.
4. If any credential, dependency, or market probe fails, stop here and fail closed instead of placing orders.

Dependency Validation

Dependency validation is required before live trading. Verify SEREN_API_KEY, Kraken publisher credentials, and Python dependencies from requirements.txt are installed and loaded. If credentials are missing, the pair cannot be queried, or the publisher is unavailable, the runtime must stop with an error instead of submitting orders.

Live Safety Opt-In

Default mode is dry-run. Live trading requires:

• python scripts/agent.py start --config config.json --allow-live
• or python scripts/agent.py cycle --config config.json --allow-live
• the normal startup risk checks to pass

The --allow-live flag is a startup-only opt-in for that process. It is not a per-order approval prompt.

Emergency Exit Path

To stop trading immediately, run python scripts/agent.py stop --config config.json. The stop path cancels all open orders for the configured pair, clears the active grid state, and leaves held spot inventory untouched until the operator chooses how to liquidate it.

SerenDB Persistence (MCP-native)

Set these optional environment variables in .env:

• SERENDB_PROJECT_NAME (default auto target: krakent)
• SERENDB_DATABASE (default auto target: krakent)
• SERENDB_BRANCH (optional)
• SERENDB_REGION (default: aws-us-east-1)
• SERENDB_AUTO_CREATE (default: true)
• SEREN_MCP_COMMAND (default: seren-mcp)

Adaptive mode requires SerenDB/MCP. If the persistence layer is unavailable, the runtime fails closed rather than falling back to local adaptive state files, runtime lock files, or local review/alert telemetry files.

Configuration

See config.example.json for available parameters including grid spacing, order size, trading pair selection, daily loss caps, cooldowns, shadow thresholds, and adaptive lock lease settings.