v1.1.0

Minimax Crypto Trading

BenignClawScan verdict for this skill. Analyzed May 1, 2026, 7:59 AM.

Analysis

This instruction-only crypto trading skill has no code, install steps, or credentials, but it gives actionable trade signals and asks for recurring review of trading performance, so users should keep human oversight.

GuidanceBefore installing, decide whether you want an agent to provide direct crypto trade signals. Do not connect it to trading automation without explicit approvals, and share only the account-risk or trade-history information you are comfortable putting in the agent context.

Findings (3)

Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.

Abnormal behavior control

Checks for instructions or behavior that redirect the agent, misuse tools, execute unexpected code, cascade across systems, exploit user trust, or continue outside the intended task.

Human-Agent Trust Exploitation

SeverityMediumConfidenceHighStatusNote

SKILL.md

A professional-grade crypto trading decision agent... outputs only EXECUTE or NO TRADE decisions

The skill presents itself as a trading decision agent and uses imperative trade labels, which users could treat as financial instructions even though no exchange execution capability is present.

User impactA user could lose money if they follow the trade signal without independent review.

RecommendationUse the skill only as decision support, verify market data independently, and require explicit human approval before placing any trade.

Rogue Agents

SeverityLowConfidenceMediumStatusNote

SKILL.md

Weekly Review Agent (AI Reviewer) — Must execute every 7 days

The skill instructs a recurring weekly review, which is disclosed and purpose-aligned but should be user-controlled if connected to scheduling or automation.

User impactIf operationalized with automation, the agent could continue producing trading reviews beyond a single user request.

RecommendationOnly enable recurring review with an explicit schedule and make sure the user can pause or disable it.

Sensitive data protection

Checks for exposed credentials, poisoned memory or context, unclear communication boundaries, or sensitive data that could leave the user's control.

Memory and Context Poisoning

SeverityLowConfidenceMediumStatusNote

SKILL.md

Risk ≤ 1% of account... Weekly Review Agent... Which pattern contributed most profit?

The workflow may involve account-size, drawdown, trade-history, and profit-pattern information, which are sensitive financial context if stored or reused by an agent.

User impactPrivate trading performance or account-risk details may become part of the agent context if the user provides them.

RecommendationProvide only the financial details needed for the analysis, avoid sharing exchange credentials, and clear or limit memory if trade history should not persist.