DingTalk Minutes

Security checks across malware telemetry and agentic risk

Overview

This skill is a local meeting-minutes organizer that reads configured transcript folders and writes local Markdown summaries, with no evidence of hidden execution or data exfiltration.

Install only if you are comfortable letting the agent read the configured meeting-record folders and save derived summaries under `./memory/meetings/`. Use a dedicated input folder for transcripts you intend to process, restrict access to the archive if meetings are confidential, and review generated or updated minutes before sharing or relying on them.

SkillSpector

By NVIDIA

Vulnerability Patterns

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
Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep

Findings (6)

Missing User Warnings

Medium

Confidence: 88% confidence
Finding: The README advertises automatic scanning of local meeting-record directories and automatic archiving, but it does not clearly warn users that the skill will read local files and create new files in memory/meetings paths. Because meeting records often contain sensitive business information, insufficient disclosure can lead users to grant access or invoke the skill without understanding the privacy and filesystem effects.

Missing User Warnings

Medium

Confidence: 92% confidence
Finding: The skill is designed to extract participants, topics, decisions, and action items from meeting transcripts, which are commonly sensitive internal or client-related data. Omitting a privacy warning increases the risk of accidental processing of confidential information and may cause users to expose regulated or proprietary data without informed consent.

Missing User Warnings

Medium

Confidence: 82% confidence
Finding: The README explicitly describes automatic scanning of local meeting-record directories and automatic archival into local memory paths, but it does not clearly warn users that the skill will read user files and create or reorganize output on disk. For a skill handling potentially sensitive meeting notes, this omission can lead to unintended data processing, silent file creation, or user surprise about where confidential content is stored.

Missing User Warnings

Medium

Confidence: 88% confidence
Finding: The skill instructs the agent to scan local directories for meeting records and generate archived files, but it does not clearly warn the user up front that it will read from and write to the filesystem. In a skill that processes potentially sensitive meeting transcripts, silent file access and persistence can expose confidential business data or surprise users with unintended local changes.

Missing User Warnings

Medium

Confidence: 91% confidence
Finding: The first-run flow creates classification folders and archives processed meeting summaries, but the skill text does not provide a clear upfront data-handling warning before these filesystem modifications occur. Because the data involved is meeting content that may contain sensitive participants, decisions, and action items, undisclosed creation of folders and stored summaries increases privacy and data-governance risk.

Missing User Warnings

Medium

Confidence: 90% confidence
Finding: The update mechanism states that archived files should be synchronized after user edits, implying modification or overwrite of existing files, but it does not warn users that existing records may be changed. This is dangerous because archived meeting minutes are business records; silent updates can cause loss of user edits, integrity issues, or unintentional alteration of important historical documentation.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal