ClawHub
Back to skill

Security audit

Slack Crawler Free

Security checks across malware telemetry and agentic risk

Overview

The skill is mostly a local Slack archive search tool, but its write-capable permissions and broad/generic instructions conflict with its read-only claims.

Review this skill before installing. Use it only for the intended local Slack archive path, avoid applying it to unrelated databases, and do not allow create, modify, reset, import, export, or save actions unless you separately understand what local files or databases will be changed.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Behavioral ASTexec() Call, eval() Call, Dynamic Import
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Findings (3)

Intent-Code Divergence

High
Confidence
93% confidence
Finding
The skill claims to provide only read-only SQL and no data modification, but the parameter descriptions advertise create/modify/import/export-style operations. In an agent context with declared write and exec tools, this inconsistency can cause the agent or user to authorize file/database writes they did not expect, undermining the trust boundary around a supposedly read-only skill.

Vague Triggers

Medium
Confidence
90% confidence
Finding
The trigger condition is overly broad: 'use when database operations, SQL queries, or data storage management are needed.' In an agent ecosystem, such broad routing language can cause the skill to activate for unrelated tasks involving local data or SQL, increasing the chance that exec/write-capable behavior is invoked unnecessarily against sensitive local resources like the Slack archive database.

Missing User Warnings

Medium
Confidence
95% confidence
Finding
The documentation advertises create/export/save/convert/reset/import-style operations without a clear warning that these may write to local files or modify local state. Because the skill also declares write and exec tools, users may believe they are using a safe read-only Slack analysis utility when the agent can in fact perform persistence-changing actions on the host filesystem.

VirusTotal

VirusTotal findings are pending for this skill version.

View on VirusTotal

Static analysis

No suspicious patterns detected.