Security audit

daily-report-tool-free

Security checks across malware telemetry and agentic risk

Overview

This daily-report skill mostly matches its stated purpose, but it asks for broad command and file-inspection powers and includes unclear API, network, and credential-check guidance for a simple local report writer.

Review before installing. This skill appears to be a daily-report generator, not malware, but it requests broad local command and search tools and gives vague API/network and credential-configuration guidance. Use it only if you are comfortable with an agent that can inspect local files and run commands; confirm the output path before use, and do not let it scan unrelated environment secrets or send report contents to callbacks or external APIs unless you explicitly intend that.

SkillSpector

By NVIDIA

Vulnerability Patterns

Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration

Findings (7)

Description-Behavior Mismatch

Medium

Confidence: 92% confidence
Finding: The skill is presented as a simple local daily-report generator, but later documentation introduces external API and network requirements. This mismatch can cause an agent or user to permit outbound connectivity and data handling that exceed the expected trust boundary, increasing the risk of unintended data exposure.

Context-Inappropriate Capability

Medium

Confidence: 95% confidence
Finding: The environment validation instructions enumerate and search for sensitive variables such as API keys, tokens, and secrets, which is not justified for a basic report-writing tool. Even with masking, this encourages credential discovery behavior and expands the agent's access to secrets unrelated to the stated task.

Context-Inappropriate Capability

Medium

Confidence: 97% confidence
Finding: Granting exec, glob, grep, and read capabilities to a simple report generator violates least-privilege principles. Broad shell access enables filesystem inspection and command execution far beyond generating a Markdown report, creating a larger attack surface if the skill is triggered unexpectedly or misused.

Intent-Code Divergence

High

Confidence: 98% confidence
Finding: The documentation claims all data remains local and is not uploaded, but elsewhere says network access and external APIs may be required. This contradictory security claim can mislead users into trusting the skill with sensitive work content under false assumptions about data locality.

Vague Triggers

Medium

Confidence: 89% confidence
Finding: The skill can be triggered from ordinary natural-language requests without a distinctive invocation boundary. Because the skill has file-writing and shell-adjacent capabilities, generic phrasing raises the risk of accidental activation during normal conversation and unintended local side effects.

Vague Triggers

Medium

Confidence: 90% confidence
Finding: The example trigger phrase is extremely generic and indistinguishable from routine user chat. In the context of a skill that writes files and has exec access, this makes unintentional activation more dangerous because benign conversation could cause state-changing actions.

Missing User Warnings

Medium

Confidence: 93% confidence
Finding: The skill documents automatic writing into the reports directory but does not provide a clear warning that local files will be created or modified. Hidden or underemphasized file mutation is risky because users may invoke the skill expecting text generation only, while the agent performs persistent changes on disk.

VirusTotal

VirusTotal findings are pending for this skill version.

View on VirusTotal

Static analysis

No suspicious patterns detected.