功能描述

AI-powered betting advisory for gaming platforms. Provides intelligent bet recommendations, risk analysis, bankroll management, and strategy optimization for crash games, dice, and casino games. Use when building automated betting bots or need AI gambling strategy guidance.

使用说明 (SKILL.md)

CLAW Betting AI Advisory

Name: CLAWDEBET
Author: eyeshaw

AI-powered betting advisory system for gaming platforms like Stake.com. Provides intelligent recommendations based on historical data analysis, risk tolerance, and proven betting strategies.

Website: https://clawde.xyz

Features

Smart Bet Recommendations: AI-analyzed optimal bet amounts and timing
Crash Game Analysis: Historical pattern analysis for crash games
Risk Management: Dynamic stop-loss and take-profit suggestions
Bankroll Protection: Kelly Criterion and flat betting strategies
Multi-Strategy Support: Martingale, Anti-Martingale, Fibonacci, D'Alembert

Quick Start

# Install the skill
clawhub install claw-betting-ai

# Configure your settings
cp config/default.json config/local.json
# Edit config/local.json with your preferences

Supported Games

Game	Analysis Method
Crash	Historical crashpoint analysis, multiplier patterns
Dice	Probability optimization, streak detection
Limbo	Target multiplier recommendations
Slide	Pattern recognition

Files Included

claw-betting-ai/
├── SKILL.md              # This file
├── config/
│   └── default.json      # Default configuration
├── scripts/
│   ├── analyze.py        # Crash history analyzer
│   └── recommend.py      # Bet recommendation engine
└── examples/
    ├── basic-usage.md    # Getting started guide
    └── strategies.md     # Strategy documentation

Configuration

See config/default.json for all available options:

strategy: betting strategy (conservative/balanced/aggressive)
bankroll: starting bankroll amount
baseBetPercent: base bet as % of bankroll
stopLoss: stop-loss threshold
takeProfit: take-profit target
maxBets: maximum bets per session

Core Strategies

1. Conservative (Low Risk)

Target: 1.5x - 2x multipliers
Win rate: ~50-65%
Max 1% bankroll per bet

2. Balanced (Medium Risk)

Target: 2x - 5x multipliers
Win rate: ~25-45%
Max 2% bankroll per bet

3. Aggressive (High Risk)

Target: 5x - 20x multipliers
Win rate: ~5-15%
Max 5% bankroll per bet

API Reference

See examples/basic-usage.md for full API documentation.

Get Recommendation

from scripts.recommend import get_recommendation

result = get_recommendation(
    bankroll=100,
    strategy="balanced",
    recent_history=[1.2, 3.5, 1.8, 2.1, 5.2]
)
# Returns: { shouldBet: true, amount: 2.0, target: 2.5, confidence: 72 }

Safety Features

Tilt Detection: Warns when betting patterns indicate emotional decisions
Session Limits: Enforces time and loss limits
Profit Locking: Auto-protects portion of winnings
Reality Checks: Periodic reminders

Disclaimer

This is an advisory system only. Gambling involves risk of loss. Never bet more than you can afford to lose. Past patterns do not guarantee future results.

Links

Website: https://clawde.xyz
Support: Contact via website

安全使用建议

This skill appears to do what it says (local analysis and bet recommendations) and does not request secrets or call external endpoints from the included code. Before installing: 1) Be aware the documentation has small inconsistencies — the example API calls omit required parameters present in the code (current_balance, session_profit, consecutive_losses). Test the functions locally to confirm the correct call signature. 2) The package contains Python scripts but does not declare Python as a required runtime or provide an installer; ensure your agent environment has a compatible Python interpreter and place files so their relative config path (config/default.json) is preserved. 3) The SKILL.md mentions a 'clawhub install' command and a website (clawde.xyz) but there is no bundled installer or verified homepage; treat those as informational only. 4) This skill advises on gambling — it does not eliminate risk; only use with caution and in jurisdictions where gambling-related automation is permitted. If you need higher assurance, run the scripts in a sandbox, review the code yourself, and confirm how the skill will be installed and invoked by your agent before granting it autonomous execution.

功能分析

Type: OpenClaw Skill Name: clawde Version: 1.0.0 The OpenClaw AgentSkills skill bundle 'clawde' is classified as benign. All files, including Python scripts and markdown documentation, are clearly aligned with the stated purpose of providing AI-powered betting advisory. The Python scripts (`analyze.py`, `recommend.py`) perform mathematical calculations based on provided input and internal configuration (`config/default.json`) without any high-risk operations such as network requests, arbitrary file system access, or command execution. The markdown files (`SKILL.md`, `examples/*.md`) contain standard installation and usage instructions for the skill and do not exhibit any prompt injection attempts or instructions for the AI agent to perform malicious actions. No evidence of data exfiltration, persistence, obfuscation, or other malicious indicators was found.

能力评估

✓ Purpose & Capability

Name/description match the included code and docs: analyze.py computes crash-game statistics and recommend.py returns bet recommendations and bankroll/risk logic. The scripts implement the advertised features (recommendations, bankroll/risk handling, multiple strategies).

ℹ Instruction Scope

Runtime instructions and examples are within the skill's scope (how to copy config, call analysis and recommendation functions). However, there's a notable documentation mismatch: several SKILL.md examples show a simplified get_recommendation(...) signature (only bankroll/strategy/history) while scripts/recommend.py require additional arguments (current_balance, session_profit, consecutive_losses). Examples also reference 'clawhub install' and a ~/.agent/skills path without an accompanying install spec or explanation of the 'clawhub' tool — this is an inconsistency in usage docs that could cause runtime confusion.

ℹ Install Mechanism

There is no install spec (instruction-only), and all code is included in the package (two Python scripts, config and docs) — no external downloads or package installs. Minor gap: the skill contains Python code but does not declare Python as a required binary/runtime; the SKILL.md suggests using 'clawhub install' even though no install mechanism is provided in the bundle.

✓ Credentials

The skill does not request environment variables, credentials, or config paths. The code reads a local config JSON file packaged with the skill; there are no network calls or references to external tokens or secrets in the source files.

✓ Persistence & Privilege

The skill is not forced-always (always:false) and uses no special system-wide privileges or modification of other skills. It does not store credentials or assert persistent elevated presence.

版本历史

v1.0.0

Initial release of CLAWDEBET. This skill provides AI-assisted betting analysis and recommendations using a strict advisory-only model. Includes session-aware analysis, bankroll management, risk controls, and strategy configuration. All recommendations are non-executing and designed for decision support only.

元数据

Slug clawde

版本 1.0.0

许可证 —

累计安装 1

当前安装数 1

历史版本数 1

常见问题

CLAWDEBET 是什么？

AI-powered betting advisory for gaming platforms. Provides intelligent bet recommendations, risk analysis, bankroll management, and strategy optimization for crash games, dice, and casino games. Use when building automated betting bots or need AI gambling strategy guidance. 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件，目前累计下载 1700 次。

如何安装 CLAWDEBET？

在 OpenClaw 或 Claude Code 对话框中运行命令「/install clawde」即可一键安装，无需额外配置。

CLAWDEBET 是免费的吗？

是的，CLAWDEBET 完全免费（开源免费），可自由下载、安装和使用。

CLAWDEBET 支持哪些平台？

CLAWDEBET 跨平台运行，可在任意部署了 OpenClaw / Claude Code 的环境中使用（cross-platform）。

谁开发了 CLAWDEBET？

由 eyeshaw（@eyeshaw）开发并维护，当前版本 v1.0.0。

CLAWDEBET