AcidDoc

• Outcome: Review
• Latest audit: May 28, 2026, 10:52 AM
• Version:
  1.0.0

Overview

This skill is for autonomous music publishing, but it defaults to recurring public submissions with wallet-linked earnings and weak approval controls.

Install only if you intentionally want an agent that can generate and publish tracks on a recurring schedule. Before running it, disable auto-submit or require manual approval, verify the exact skill identifier and publisher, avoid remote skill URLs you have not inspected, use limited-scope API keys and account tokens, keep wallet withdrawal disabled or isolated to a low-risk wallet, and monitor posted tracks, API usage, and earnings activity.

SkillSpector

By NVIDIA

Vulnerability Patterns

• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• Excessive Agency: Unrestricted Tool Access, Autonomous Decision Making, Scope Creep
• Trigger Abuse: Overly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands
• Privilege Escalation: Excessive Permissions, Sudo/Root Execution, Credential Access

Findings (9)

Missing User Warnings

Medium

• Confidence: 90% confidence
• Finding: The guide explicitly configures `--auto-submit true`, which enables autonomous external actions on the user's behalf without a prominent warning or explicit consent checkpoint. Because this skill submits generated content to an external platform and is framed as a production workflow, users may unknowingly authorize ongoing account-affecting actions, spammy behavior, or undesired public submissions.

Missing User Warnings

Medium

• Confidence: 87% confidence
• Finding: The troubleshooting/setup flow instructs users to connect a wallet and operate an autonomous production agent without clearly disclosing the financial and account-impacting implications. Even if the wallet is only used for platform interactions, normalizing wallet connection in an unattended agent increases the risk of unintended transactions, asset exposure, or misuse if the environment is compromised.

Missing User Warnings

Medium

• Confidence: 93% confidence
• Finding: The README advertises an automated workflow that composes content, submits it to claw.fm, and collects USDC, but it does not clearly warn users that the skill may transmit generated content and metadata to external services on a recurring basis. In a skill that performs autonomous posting and monetized activity, omission of disclosure about outbound network actions and external publication can lead to unintended data sharing, account misuse, or reputational harm.

Missing User Warnings

Medium

• Confidence: 91% confidence
• Finding: The README promotes USDC earnings and references wallet/API key usage without prominently warning that users may be linking financial endpoints and paid third-party accounts to an autonomous agent. In this context, missing disclosure increases the risk of users exposing credentials, incurring charges, or enabling unintended financial actions without informed consent.

Missing User Warnings

Medium

• Confidence: 95% confidence
• Finding: The skill explicitly promotes autonomous track submission, wallet connection, earnings monitoring, and possible future fund flows, but it does not present clear upfront warnings or consent boundaries around financial and publishing actions. In an agent context, that can cause users to enable automatic external actions without understanding that content may be published publicly and linked to a wallet, creating monetary, reputational, and operational risk.

Missing User Warnings

Medium

• Confidence: 91% confidence
• Finding: The analytics section references geographic distribution and listener demographics without any privacy notice, data minimization statement, or explanation of lawful/appropriate use. Even if the skill only surfaces platform-provided aggregates, presenting audience profiling features without privacy guidance can encourage collection or misuse of potentially sensitive listener data.

Missing User Warnings

Medium

• Confidence: 94% confidence
• Finding: The skill explicitly promotes autonomous submission of generated tracks to an external service and links this behavior to wallet-based earnings, but it does not provide a clear, upfront warning about ongoing network transmission of audio, metadata, and account-linked activity. In an agent context, repeated autonomous uploads can expose user data, operational metadata, and financial identifiers without sufficiently informed consent.

Missing User Warnings

Medium

• Confidence: 95% confidence
• Finding: The skill instructs users to supply third-party music-generation API keys and use external providers, but it does not clearly warn that prompts, generated content, and possibly related metadata will be sent to those providers. This creates a transparency and data-handling risk, especially where users may assume generation happens locally or may not understand that multiple vendors receive content.

Vague Triggers

Medium

• Confidence: 90% confidence
• Finding: The configuration explicitly instructs the agent to autonomously produce and submit tracks every 12 hours, and also sets auto_submit to true in multiple sections. This creates unattended external actions without clear approval gates, failure conditions, rate limits beyond the basic schedule, or safety checks, which can lead to spam, unintended publishing, or abuse if the agent behavior drifts or dependencies are compromised.

VirusTotal

64/64 vendors flagged this skill as clean.