← Back to Skills Marketplace
richducat

Prediction Trade Journal

by richducat · GitHub ↗ · v1.0.0 · MIT-0
cross-platform ⚠ suspicious
87
Downloads
0
Stars
0
Active Installs
1
Versions
Install in OpenClaw
/install dolph-trade-journal
Description
Auto-log trades with context, track outcomes, generate calibration reports to improve trading.
README (SKILL.md)

Prediction Trade Journal

Track every trade, learn from outcomes, improve your edge.

When to Use This Skill

Use this skill when the user wants to:

  • See their trade history
  • Track win rate and P&L
  • Generate trading reports
  • Analyze which strategies work best

Quick Commands

# Sync trades from API
python tradejournal.py --sync

# Show recent trades
python tradejournal.py --history 10

# Generate weekly report
python tradejournal.py --report weekly

# Export to CSV
python tradejournal.py --export trades.csv

API Reference:

  • Base URL: https://api.simmer.markets
  • Auth: Authorization: Bearer $SIMMER_API_KEY
  • Trades: GET /api/sdk/trades

How It Works

  1. Sync - Polls /api/sdk/trades to fetch trade history
  2. Store - Saves trades locally with outcome tracking
  3. Track - Updates outcomes when markets resolve
  4. Report - Generates win rate, P&L, and calibration analysis

CLI Reference

Command Description
--sync Fetch new trades from API
--history N Show last N trades (default: 10)
--sync-outcomes Update resolved markets
--report daily/weekly/monthly Generate summary report
--config Show configuration
--export FILE.csv Export to CSV
--dry-run Preview without making changes

Configuration

Setting Environment Variable Default
API Key SIMMER_API_KEY (required)

Storage

Trades are stored locally in data/trades.json:

{
  "trades": [{
    "id": "uuid",
    "market_question": "Will X happen?",
    "side": "yes",
    "shares": 10.5,
    "cost": 6.83,
    "outcome": {
      "resolved": false,
      "winning_side": null,
      "pnl_usd": null
    }
  }],
  "metadata": {
    "last_sync": "2025-01-29T...",
    "total_trades": 50
  }
}

Skill Integration

Other skills can enrich trades with context:

from tradejournal import log_trade

# After executing a trade
log_trade(
    trade_id=result['trade_id'],
    source="copytrading",
    thesis="Mirroring whale 0x123...",
    confidence=0.70
)

This adds thesis, confidence, and source to the trade record for better analysis.

Example Report

📓 Weekly Report
========================================
Period: Last 7 days
Trades: 15
Total cost: $125.50
Resolved: 8 / 15
Win rate: 62.5%
P&L: +$18.30

By side: 10 YES, 5 NO

Troubleshooting

"SIMMER_API_KEY environment variable not set"

  • Set your API key: export SIMMER_API_KEY=sk_live_...

"No trades recorded yet"

  • Run python tradejournal.py --sync to fetch trades from API

Trades not showing outcomes

  • Run python tradejournal.py --sync-outcomes to update resolved markets
Usage Guidance
This package largely does what it says: it syncs trades from Simmer and saves them locally. Before installing, verify the following: 1) Confirm you trust the 'simmer-sdk' PyPI package (clawhub.json lists it) — inspect that package or install in an isolated env. 2) Protect your SIMMER_API_KEY: the skill will use it to fetch your trades; consider creating a read-only or limited API key if Simmer supports it. 3) Be aware data/trades.json and context.json are stored in the skill folder and may contain sensitive trade positions/thesis — treat or encrypt/limit access accordingly. 4) Note metadata/documentation mismatches (registry header vs. clawhub.json vs. SKILL.md): the code also uses SIMMER_API_URL and SIMMER_JOURNAL_* env vars even if not documented. 5) If you want stronger assurance, review the full tradejournal.py and the simmer-sdk dependency source, or run the skill in an isolated container/VM and monitor outbound requests (they should go to api.simmer.markets or your SIMMER_API_URL).
Capability Analysis
Type: OpenClaw Skill Name: dolph-trade-journal Version: 1.0.0 The skill bundle is a legitimate trade journaling tool designed to sync, track, and report on trading activity from the Simmer Markets API. The code in tradejournal.py uses standard libraries (urllib) to interact with its stated API endpoint (api.simmer.markets) and manages local data storage in JSON format. There is no evidence of malicious intent, data exfiltration to unauthorized domains, or prompt injection attempts in the documentation or scripts.
Capability Tags
cryptocan-make-purchases
Capability Assessment
Purpose & Capability
Name/description describe a trade journal and the code implements exactly that: polling https://api.simmer.markets, storing trades in data/trades.json, updating outcomes, exporting reports. Requesting an API key (SIMMER_API_KEY) is appropriate for this purpose. However, registry-level 'Requirements' listed at the top of the pack show none while clawhub.json and SKILL.md require SIMMER_API_KEY and a pip dependency — a metadata mismatch.
Instruction Scope
SKILL.md instructions map to the code: sync, history, report, export. The runtime instructions and code only access the Simmer API and local files in the skill directory (data/, config.json). The code does read environment variables beyond the single documented SIMMER_API_KEY (it also honors SIMMER_API_URL and SIMMER_JOURNAL_* env vars) and exposes a log_trade() integration point for other skills to add context to local records — this is expected but worth noting.
Install Mechanism
This is instruction+code with no explicit install spec, which is low-risk. However clawhub.json declares a pip dependency ('simmer-sdk') even though there's no install script in the package metadata — installing that dependency would pull third-party code from PyPI. That is a standard dependency for this purpose but should be trusted/reviewed before installation.
Credentials
The skill requires a single service credential (SIMMER_API_KEY) which is proportional to fetching trades. Additional environment variables (SIMMER_API_URL, SIMMER_JOURNAL_FETCH_LIMIT, SIMMER_JOURNAL_AUTO_SYNC) are used by the code but not all are documented in SKILL.md or clawhub.json; this is sloppy but not necessarily malicious. The skill does not request unrelated secrets or cloud credentials.
Persistence & Privilege
The skill stores data and config under its own directory (data/*.json, config.json) and does not request always:true or elevated system privileges. It can be invoked autonomously (normal default) but does not modify other skills or system-wide settings.
How to Use
  1. Make sure OpenClaw is installed (local or Docker)
  2. Run the install command in chat: /install dolph-trade-journal
  3. After installation, invoke the skill by name or use /dolph-trade-journal
  4. Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.0
- Initial release of the Prediction Trade Journal skill (version 1.0.0). - Automatically logs trades, tracks their context and outcomes, and stores them locally. - Provides CLI commands for syncing trades, viewing history, generating reports, and exporting to CSV. - Supports configuration with a required API key (SIMMER_API_KEY). - Allows skill integration to enrich trade logs with thesis, confidence, and source. - Generates reports with win rate, P&L, and calibration analysis to help improve trading.
Metadata
Slug dolph-trade-journal
Version 1.0.0
License MIT-0
All-time Installs 0
Active Installs 0
Total Versions 1
Frequently Asked Questions

What is Prediction Trade Journal?

Auto-log trades with context, track outcomes, generate calibration reports to improve trading. It is an AI Agent Skill for Claude Code / OpenClaw, with 87 downloads so far.

How do I install Prediction Trade Journal?

Run "/install dolph-trade-journal" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.

Is Prediction Trade Journal free?

Yes, Prediction Trade Journal is completely free, licensed under MIT-0. You can download, install and use it at no cost.

Which platforms does Prediction Trade Journal support?

Prediction Trade Journal is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).

Who created Prediction Trade Journal?

It is built and maintained by richducat (@richducat); the current version is v1.0.0.

💬 Comments