Security audit

AI Meeting Intelligence Expert

Security checks for vulnerabilities and agentic risk

Overview

This is a plain Markdown meeting-assistant skill whose behavior is purpose-aligned and includes sensible privacy and confirmation guidance.

Before installing, expect this skill to handle potentially sensitive meeting content. Use its own guidance: redact personal, financial, customer, medical, and confidential details where possible; get participant consent for recording or transcription; and review any draft carefully before saving, sending, or syncing it to other tools.

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 (1)

Vague Triggers

Medium

Confidence: 87% confidence
Finding: The trigger list includes broad phrases such as "action items", "meeting notes", and "meeting summary" without clear activation boundaries or exclusion conditions. These are common everyday work phrases and could cause unintended invocation when users discuss normal meeting artifacts rather than explicitly requesting this skill.

Static analysis

No suspicious patterns detected.