Security audit

Slack Integration

Security checks across malware telemetry and agentic risk

Overview

This is a coherent Slack API integration, but users should grant narrow Slack scopes and handle bot tokens carefully.

Install only if you trust the Slack workspace impact. Grant only the scopes needed for the features you will use, avoid broad search/private-channel/DM scopes unless necessary, store the bot token securely, never paste real tokens into chats or tracked files, and confirm local files or URL content before uploading them to Slack.

SkillSpector

By NVIDIA

Vulnerability Patterns

Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code

Findings (5)

Intent-Code Divergence

Medium

Confidence: 98% confidence
Finding: The installation guide includes a realistic-looking Slack bot token even though the document later warns not to expose tokens. In a skill that teaches users to configure live Slack access, publishing credential-shaped secrets can lead to copy-paste misuse, accidental deployment of exposed credentials, or normalization of unsafe secret-handling practices.

Missing User Warnings

Medium

Confidence: 90% confidence
Finding: The skill enables listing channels, searching messages, and looking up user details without warning that these actions expose potentially sensitive workspace metadata and personal information. In a Slack integration context, this increases the risk of privacy-invasive use and overcollection because operators may not realize the breadth of data being queried and returned.

Missing User Warnings

Medium

Confidence: 95% confidence
Finding: The file upload capability states that local paths or URLs can be uploaded to Slack but does not warn that this transmits local or fetched content to an external third party. In an agent setting, this omission can cause accidental exfiltration of sensitive local files or remote content that the user did not intend to share externally.

Missing User Warnings

Low

Confidence: 93% confidence
Finding: The installation guide includes realistic secret values without warning readers not to paste actual credentials into shared configuration examples or documentation. While this is partly documentation hygiene, it materially increases the chance of accidental secret disclosure in copied config files, tickets, repos, or screenshots.

Ssd 3

High

Confidence: 99% confidence
Finding: The example configuration exposes realistic plaintext values for a Slack bot token and signing secret. In a security-sensitive integration skill, embedding credential-like values is dangerous because they may be mistaken for real secrets, reused insecurely, scraped automatically, or train users to handle secrets unsafely.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.