ClawHub
Back to skill

Security audit

bom-vuln-intel-tool-free

Security checks across malware telemetry and agentic risk

Overview

The skill mostly matches an SBOM and dependency-vulnerability purpose, but its trigger text incorrectly tells agents to use it for database and SQL work while it has read, exec, and networked lookup behavior.

Review the trigger mismatch before installing or using this skill. It should only be invoked for SBOM, package inventory, and dependency vulnerability checks, and users should avoid running its networked lookups on private/internal dependency names unless sending package metadata to npm, PyPI, or OSV is acceptable.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (7)

Intent-Code Divergence

High
Confidence
97% confidence
Finding
The documented trigger condition says the skill should be used for database operations, SQL queries, and data storage management, which directly contradicts the skill's actual SBOM and dependency-vulnerability purpose. In an agent environment, this mismatch can cause the skill to activate in unrelated contexts and run networked or shell-based actions against the wrong project or user intent.

Vague Triggers

High
Confidence
98% confidence
Finding
The trigger-condition section is obviously inconsistent with the documented capabilities, creating a prompt-routing and tool-selection hazard. This can lead an agent to invoke this skill during unrelated database tasks, exposing local project metadata or initiating external requests under false assumptions.

Missing User Warnings

Medium
Confidence
93% confidence
Finding
The skill repeatedly instructs the agent to send package names, versions, and potentially inferred dependency inventories to external services such as npm, PyPI, and OSV, but does not present a clear privacy or data-egress warning to the user. In some environments, dependency names, internal package identifiers, or project metadata can be sensitive and their transmission may violate policy or leak information about the codebase.

External Transmission

Medium
Category
Data Exfiltration
Content
# ...
    echo "检查: ${pkg}@${version} (${ecosystem})"
# ...
    RESULT=$(curl -s -X POST "https://api.osv.dev/v1/query" \
        -H "Content-Type: application/json" \
        -d "{
            \"package\": {
Confidence
94% confidence
Finding
curl -s -X POST "https://api.osv.dev/v1/query" \ -H "Content-Type: application/json" \ -d

External Transmission

Medium
Category
Data Exfiltration
Content
VER=$(curl -s "https://registry.npmjs.org/${PACKAGE}" | jq -r '."dist-tags".latest')
fi
# ...
curl -s -X POST "https://api.osv.dev/v1/query" \
    -H "Content-Type: application/json" \
    -d "{\"package\": {\"name\": \"${PACKAGE}\", \"ecosystem\": \"npm\"}, \"version\": \"${VER}\"}" | \
    jq '.vulns | if . then "发现 \(length) 个漏洞" else "未发现已知漏洞" end'
Confidence
94% confidence
Finding
curl -s -X POST "https://api.osv.dev/v1/query" \ -H "Content-Type: application/json" \ -d

External Transmission

Medium
Category
Data Exfiltration
Content
# ...
    echo "检查: ${pkg}@${version} (${ecosystem})"
# ...
    RESULT=$(curl -s -X POST "https://api.osv.dev/v1/query" \
        -H "Content-Type: application/json" \
        -d "{
            \"package\": {
Confidence
94% confidence
Finding
https://api.osv.dev/

External Transmission

Medium
Category
Data Exfiltration
Content
VER=$(curl -s "https://registry.npmjs.org/${PACKAGE}" | jq -r '."dist-tags".latest')
fi
# ...
curl -s -X POST "https://api.osv.dev/v1/query" \
    -H "Content-Type: application/json" \
    -d "{\"package\": {\"name\": \"${PACKAGE}\", \"ecosystem\": \"npm\"}, \"version\": \"${VER}\"}" | \
    jq '.vulns | if . then "发现 \(length) 个漏洞" else "未发现已知漏洞" end'
Confidence
94% confidence
Finding
https://api.osv.dev/

VirusTotal

VirusTotal findings are pending for this skill version.

View on VirusTotal

Static analysis

No suspicious patterns detected.