Security audit

AutoWhisper

Security checks across malware telemetry and agentic risk

Overview

AutoWhisper is a disclosed marketing automation connector that can use an API token to create, schedule, and publish social content for users who intentionally grant it that access.

Install this only if you want an agent to work with AutoWhisper on your behalf, including sending product information to AutoWhisper and publishing to social accounts you connect. Keep the API token private, consider restricting the credential file permissions, review posts before publishing unless you intentionally enable autonomous operation, and rotate or revoke the token if the machine or workspace is shared.

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 (5)

Context-Inappropriate Capability

Medium

Confidence: 90% confidence
Finding: The skill instructs the agent to read and use a bearer token from a local credentials file, enabling access to an external service with account-scoped privileges. While token use is expected for API-backed skills, the instructions do not constrain scope, require explicit per-action user consent, or warn about secure handling, so a broadly-triggered skill could access and act on behalf of the user more often than intended.

Missing User Warnings

Medium

Confidence: 93% confidence
Finding: The README explicitly tells users that after connecting a social account, the agent can publish on its own, but it does not warn that this is an account-impacting action or recommend confirmation gates. In an agentic context, that omission increases the chance of unintended posts, brand damage, or abuse if the agent is prompted incorrectly or credentials are misused.

Missing User Warnings

Medium

Confidence: 88% confidence
Finding: The README says the agent sends product and competitor-related information to AutoWhisper over an HTTP API, but it does not clearly disclose what data may be transmitted, stored, or processed. In a marketing skill that learns from user-provided product data and competitor activity, this can lead to unintentional sharing of sensitive business information without informed consent.

Vague Triggers

High

Confidence: 95% confidence
Finding: The activation criteria are extremely broad: the skill should be used whenever the user asks to run marketing, generate content for a product, or merely mentions AutoWhisper. In combination with credentialed API access and action-taking capabilities, this increases the chance of unintended invocation, causing the agent to contact the external service or take account actions without sufficiently specific user intent.

Missing User Warnings

Medium

Confidence: 93% confidence
Finding: The skill tells users to persist an API token in a plaintext file under the home directory and provides no warning about local secret exposure, file permissions, or rotation. This raises the risk of credential leakage to other tools, users, logs, backups, or misconfigured environments, especially because the token grants direct access to account operations.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.