Back to skill

Security audit

BlueColumn ClawHub Integration

Security checks across malware telemetry and agentic risk

Overview

The skill is a coherent BlueColumn audio integration, but it under-discloses sensitive audio storage and permits insecure API-key storage in a markdown file.

Review before installing. Use the platform secret store only, not TOOLS.md, for the BlueColumn API key. Treat all uploaded audio, transcripts, summaries, entities, sentiment, embeddings, identifiers, and webhook payloads as sensitive; enable this only with user consent, clear retention expectations, and secured webhook endpoints.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (4)

Context-Inappropriate Capability

Medium
Confidence
99% confidence
Finding
The skill explicitly permits storing a live API key in TOOLS.md, which is typically documentation or repository content rather than a protected secret store. That guidance materially increases the chance of credential leakage through version control, logs, sharing, or downstream agent/tool exposure, enabling unauthorized access to BlueColumn data and services.

Missing User Warnings

Medium
Confidence
90% confidence
Finding
The skill describes sending user audio, identifiers, and metadata to a third-party API without an explicit privacy notice, consent requirement, or data-handling warning. Because audio content may contain sensitive personal, financial, or health information, omission of these warnings can lead to silent exfiltration of regulated or confidential data.

Missing User Warnings

Medium
Confidence
94% confidence
Finding
The default configuration enables store_original_audio without explaining retention, access control, or privacy consequences. Persisting raw audio substantially raises the sensitivity of the integration because the original recording may contain unredacted voiceprints and confidential content beyond derived transcripts or summaries.

Missing User Warnings

Low
Confidence
85% confidence
Finding
The webhook guidance omits that callbacks may contain transcripts, summaries, entities, sentiment, or other sensitive derived data, and that insecure callback endpoints could leak this information. Without validation and transport safeguards, attackers or misconfigured systems may intercept or spoof processing results.

VirusTotal

VirusTotal findings are pending for this skill version.

View on VirusTotal

Static analysis

Detected: suspicious.exposed_secret_literal

File appears to expose a hardcoded API secret or token.

Critical
Code
suspicious.exposed_secret_literal
Location
SKILL.md:124