Security audit

Meeting Minutes Workflow

Security checks for vulnerabilities and agentic risk

Overview

The skill is for meeting minutes, but its Feishu sync command is unsafe and it can share or store meeting content without clear consent.

Install only if you are comfortable reviewing each Feishu sync and local action-item save before it happens. Do not run the documented Feishu shell command with untrusted meeting text; use a safer argument-array or file-based workflow instead.

A.I.G

Vulnerability Patterns

Insecure Skill Coding PracticesFinds exploitable flaws such as hardcoded secrets or command injection
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

Findings (1)

T09 · Insecure Skill Coding Practices

Error

Location: SKILL.md:31
Finding: Shell Command Injection Through Untrusted Meeting Content in Feishu Sync Workflow
Content: View full analysis
Remediation: View remediation

SkillSpector

Vulnerability Patterns

Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands

Findings (5)

Tp4

High

Category: MCP Tool Poisoning
Confidence: 98% confidence
Finding: The code partially aligns with the 'create meeting minutes templates/customize meeting formats' portion of the description because it renders a meeting-minutes template with provided values. However, it does not implement several prominent declared capabilities: it does not parse discussion notes, does not generate structured minutes from raw notes, does not sync anything to Feishu, and does not manage or track action items. Its actual purpose is much narrower: template rendering for meeting minutes. Therefore the description materially overstates the implemented behavior.
Content: No source excerpt is available for this finding.

Undeclared Tool Scope

Medium

Category: MCP Least Privilege
Confidence: 91% confidence
Finding: The skill instructs the agent to read a template file and write action-item data to local memory, but it declares no explicit tool scope or permissions. This creates an authorization gap where file access may occur without clear user or platform constraints, increasing the chance of unintended file reads/writes or future expansion of the skill beyond its stated boundaries.
Content: No source excerpt is available for this finding.

Vague Triggers

Medium

Category: Not specified by scanner
Confidence: 94% confidence
Finding: The invocation criteria are broad enough to trigger on generic meeting-related context, which can cause the agent to activate this skill when the user did not explicitly request note generation, syncing, or tracking. In this skill's context, over-triggering is more dangerous because it can lead to unintended processing of sensitive meeting content and downstream storage or external sharing.
Content: No source excerpt is available for this finding.

Missing User Warnings

Medium

Category: Not specified by scanner
Confidence: 96% confidence
Finding: The Feishu sync section encourages creating a document and sharing it with attendees, but it does not require a user-facing consent step or warn that potentially sensitive meeting content will be sent to an external service and possibly shared. In the context of meeting minutes, this is especially risky because notes often contain confidential business discussions, personnel matters, or strategic decisions.
Content: No source excerpt is available for this finding.

Missing User Warnings

Low

Category: Not specified by scanner
Confidence: 94% confidence
Finding: The skill saves extracted action items to a local memory file without warning the user that meeting-derived data will be persisted. Because action items may include names, deadlines, internal projects, or sensitive follow-up tasks, silent local retention creates privacy and data-governance risks.
Content: No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.