Security audit

Credit Mastery

Security checks across malware telemetry and agentic risk

Overview

This skill is really a Swarms AI orchestration guide, not a credit skill, and it includes high-risk wallet-private-key and autonomous-agent instructions that need careful review.

Install only if you specifically intend to use Swarms AI. Treat the Credit Mastery name mismatch as a warning sign, do not paste a main Solana wallet private key into prompts or generated code, use testnet or a dedicated low-balance wallet, and require explicit approval before token launches, payments, marketplace publishing, autonomous sub-agents, or file-operation workflows.

SkillSpector

By NVIDIA

Vulnerability Patterns

Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access

Findings (7)

Context-Inappropriate Capability

Medium

Confidence: 89% confidence
Finding: The documentation states that when autonomous looping is enabled, sub-agents may access broad file-operation tools including create, update, read, list, and delete, even though the page is specifically about sub-agent delegation. In a multi-agent orchestration context, this expands the attack surface significantly: delegated agents can modify local state or exfiltrate data if prompted maliciously or if task boundaries are unclear.

Vague Triggers

Medium

Confidence: 88% confidence
Finding: The manifest description is extremely broad and covers many generic agent-building, orchestration, streaming, marketplace, payment, and token-launch scenarios. In a skill-routing system, this can cause over-activation on loosely related prompts, increasing the chance the skill is invoked in contexts where users did not intend external API calls, delegation, or blockchain-related actions.

Missing User Warnings

High

Confidence: 98% confidence
Finding: The token-launch example instructs users to include a Solana wallet private key in a request payload sent to a remote service, while only briefly noting approximate cost. Transmitting raw private keys to an external API is highly sensitive and can lead to irreversible asset theft, unauthorized signing, and unintended cost-incurring blockchain actions if the example is copied into production workflows.

Missing User Warnings

High

Confidence: 99% confidence
Finding: The documentation explicitly instructs clients to send a wallet private key in an HTTP header, which is an extremely sensitive credential transmission pattern. Even over TLS, headers are commonly exposed to reverse proxies, middleware, access logs, tracing systems, browser tooling, and server-side observability pipelines; compromise of this key enables theft of blockchain funds and irreversible transaction signing.

Missing User Warnings

High

Confidence: 98% confidence
Finding: The client example normalizes passing a raw wallet private key directly into application code without strong warnings, encouraging unsafe secret handling by downstream users. This increases the chance the key will be hardcoded, copied into source control, leaked through debugging, crash dumps, memory inspection, or reused in insecure automation, leading to wallet compromise.

Missing User Warnings

Medium

Confidence: 92% confidence
Finding: The documentation instructs users to send a Solana private key and API key to a remote API without prominent warnings about credential sensitivity, secure handling, rotation, or use of least-privilege alternatives. In an agent-building skill context, users may copy real production secrets into example requests, increasing the risk of wallet compromise, unauthorized API use, and accidental logging or exposure through tooling.

Missing User Warnings

Medium

Confidence: 94% confidence
Finding: The page documents autonomous file creation, modification, reading, listing, and deletion without any accompanying warning about local data integrity, privacy, or accidental destructive behavior. Because these tools are available in an autonomous loop and can be combined with delegated sub-agents, users may unknowingly enable agents to alter or remove files without sufficient oversight.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.