Security audit

AI Daily Briefing

Security checks across malware telemetry and agentic risk

Overview

This is a coherent daily briefing skill that reads expected productivity sources, though users should be aware it can surface private tasks, meeting notes, memory, and calendar details.

Install only in workspaces where you are comfortable having todo.md, meeting-notes, MEMORY.md, USER.md, and connected calendar details summarized back to you. Prefer explicit requests like "daily briefing" in private contexts, and ignore the current crypto/purchase capability tags unless the publisher corrects or explains them.

SkillSpector

By NVIDIA

Vulnerability Patterns

Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code

Findings (7)

Vague Triggers

Medium

Confidence: 92% confidence
Finding: The trigger phrases include broad, natural-language expressions such as "what do I need to know?", "start my day", and "give me the rundown," which are likely to appear in ordinary conversation unrelated to this skill. In systems that auto-activate skills based on phrase matching, this can cause unintended invocation, leading the agent to access and summarize workspace files like todo.md, meeting notes, memory files, or calendar data without the user explicitly requesting this skill.

Vague Triggers

Medium

Confidence: 90% confidence
Finding: The trigger phrase "briefing" is extremely broad and likely to appear in normal conversation, which can cause unintended skill activation and disclosure of sensitive task, calendar, or meeting-note data. In a productivity assistant that aggregates personal context automatically, weak activation boundaries materially increase privacy risk.

Vague Triggers

Medium

Confidence: 86% confidence
Finding: The example phrase "what's today look like?" is ordinary conversational language and could match many benign interactions unrelated to this skill. If the skill activates on such a broad phrase, it may expose private schedule and task information when the user did not clearly intend to invoke the briefing feature.

Vague Triggers

Medium

Confidence: 89% confidence
Finding: The trigger phrase "what's overdue?" is common language that may be used in many contexts, but here it could cause the skill to reveal overdue tasks and related business information. Because the skill surfaces potentially sensitive work items, ambiguous activation increases the chance of accidental disclosure in shared or misinterpreted contexts.

Vague Triggers

Medium

Confidence: 87% confidence
Finding: The example trigger phrases are broad natural-language requests like "start my day" and "give me the rundown," which could overlap with ordinary user conversation and cause the skill to activate when the user did not explicitly intend it. In a briefing skill that may surface calendar items, overdue tasks, and meeting context, unintended invocation can expose sensitive productivity data in the wrong context.

Vague Triggers

Medium

Confidence: 82% confidence
Finding: The quick-start instruction "Just say: briefing" and surrounding phrasing do not clarify how activation is scoped, whether exact matches are required, or how the agent distinguishes intentional invocation from normal dialogue. That ambiguity increases the chance of accidental activation and unintended disclosure of personal work context pulled from connected tools.

Missing User Warnings

Medium

Confidence: 90% confidence
Finding: The documentation states that meeting-derived action items and calendar data are automatically incorporated into briefings, but it gives no explicit warning that personal schedule details, task lists, and meeting context may be surfaced. Users may not realize the scope of data aggregation, which can lead to unexpected exposure of sensitive information during a briefing.

VirusTotal

48/48 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.