Security audit

CLAWDEBET

Security checks for vulnerabilities and agentic risk

Overview

This gambling advice skill is disclosed as betting-related, but it overstates its capabilities and advertises safety controls that the code does not actually enforce.

Review this carefully before installing. Treat it as simple gambling decision support, not a predictive AI system, and do not connect it to automated real-money betting unless you add independent hard limits, user confirmation for each wager, jurisdiction and age checks, and responsible-gambling controls. Do not rely on the advertised safety features as enforced by the included code.

A.I.G

Vulnerability Patterns

Insecure Skill Coding PracticesFinds exploitable flaws such as hardcoded secrets or command injection
Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
Agent Memory PoisoningWrites attacker-controlled rules into memory that affect later sessions
Remote Payload Retrieval and ExecutionFetches external code whose behavior can change after review
Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally

Findings (1)

T09 · Insecure Skill Coding Practices

Warning

Location: scripts/recommend.py:54
Finding: Documented Betting Safety Controls Are Not Enforced
Content: View full analysis
BetRecommendation: ``` Only immediate balance thresholds and a consecutive-loss cooldown decision are implemented: ```python # Check stop-loss stop_loss_threshold = bankroll * (1 - config["stopLossPercent"] / 100) if current_balance <= stop_loss_threshold: return BetRecommendation( should_bet=False, amount=0, target_multiplier=0, confidence=100, risk_level="stop", reasoning="Stop-loss triggered. Session should end.", stop_loss_hit=True ) # Check take-profit take_profit_threshold = bankroll * (1 + config["takeProfitPercent"] / 100) if current_balance >= take_profit_threshold: return ...[truncated 3114 chars]
Remediation: View remediation

SkillSpector

Vulnerability Patterns

Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code

Findings (4)

Tp4

High

Category: MCP Tool Poisoning
Confidence: 92% confidence
Finding: The skill materially overstates its capabilities by advertising AI-powered, multi-game betting advice, bankroll management, and strategy optimization while the underlying implementation reportedly only performs limited crash-history heuristics. In a gambling context, this can mislead users or downstream agents into making financial decisions based on inaccurate assumptions about sophistication, coverage, and safety controls.
Content: No source excerpt is available for this finding.

Missing User Warnings

Medium

Category: Not specified by scanner
Confidence: 92% confidence
Finding: The guide provides concrete betting recommendations, confidence scores, and strategy framing without a prominent warning that users can lose real money and that past game history does not reliably predict future outcomes. In a skill explicitly marketed for automated betting bots, this can encourage overreliance on the system and normalize risky gambling behavior.
Content: No source excerpt is available for this finding.

Missing User Warnings

Medium

Category: Not specified by scanner
Confidence: 95% confidence
Finding: The session-flow example demonstrates an open-ended betting loop that repeatedly requests recommendations and places bets, but it lacks explicit warnings about compulsive use, addiction risk, or bankroll depletion. Because the skill is intended for automated betting workflows, this pattern materially increases the chance of sustained, impulsive, or unattended gambling losses.
Content: No source excerpt is available for this finding.

Missing User Warnings

Medium

Category: Not specified by scanner
Confidence: 95% confidence
Finding: This is a true safety issue: the document provides detailed gambling strategies, progression systems, recovery advice, and bankroll guidance in a way that can encourage sustained or escalated betting behavior, while only briefly acknowledging risk at the very end. The skill context makes this more dangerous because it is explicitly marketed for automated betting bots and strategy optimization, which can normalize harmful gambling behavior and financial loss without prominent upfront harm-reduction messaging.
Content: No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.