Security audit

Daily Briefing

Security checks for vulnerabilities and agentic risk

Overview

This skill appears purpose-aligned, but it can gather sensitive calendar, email, task, note, and CRM details through broad triggers and optional automatic runs without enough scoping.

Review this before installing if your agent has access to email, calendar, CRM, tasks, or notes. Use it only with explicit source permissions, avoid broad automatic delivery unless the destination is private, and prefer narrower invocation such as a named daily briefing command.

A.I.G

Vulnerability Patterns

Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
Agent Memory PoisoningWrites attacker-controlled rules into memory that affect later sessions
Remote Payload Retrieval and ExecutionFetches external code whose behavior can change after review
Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally
Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs

SkillSpector

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 (4)

Vague Triggers

Medium

Confidence: 97% confidence
Finding: The usage example uses the extremely generic trigger phrase "Brief me.", which can be invoked accidentally in normal conversation or collide with other assistant capabilities. In a skill that accesses calendar, inbox, and news sources, unintended invocation can expose or summarize personal data without the user deliberately requesting this specific skill.

Missing User Warnings

Medium

Confidence: 95% confidence
Finding: The README encourages automatic delivery via a morning cron job but does not warn that the skill reads sensitive personal data sources such as calendar events and urgent emails. This increases the risk of users enabling unattended runs without understanding the privacy implications, which can lead to exposure of confidential schedule or inbox content in insecure channels or contexts.

Vague Triggers

Medium

Confidence: 95% confidence
Finding: The trigger phrases are broad, natural-language requests like "What's on today?" and "Start my day," which can plausibly be said in ordinary conversation and may cause unintended activation. Because this skill aggregates calendar, email, CRM, tasks, news, and weather, accidental invocation could expose sensitive summaries or cause data retrieval the user did not explicitly intend.

Missing User Warnings

Medium

Confidence: 92% confidence
Finding: The skill explicitly pulls from sensitive sources including calendar, email, CRM, tasks, and notes, but it provides no privacy notice, consent flow, or data-minimization guidance. In practice, this can lead to over-collection and over-disclosure of personal or business-sensitive information in a generated briefing, especially if displayed in shared environments or triggered automatically.

Static analysis

No suspicious patterns detected.