ClawHub
Back to skill

Security audit

GraphQL

Security checks across malware telemetry and agentic risk

Overview

This is a coherent GraphQL guidance skill with narrow, disclosed local preference storage and no executable payload, but users should sanitize debugging data before logging it.

Before installing, be aware that the skill may remember GraphQL preferences and project context in ~/Clawic/data/graphql/. When using its debugging steps, redact tokens, cookies, passwords, personal data, and tenant-specific records before sharing or storing logs.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • 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 (3)

Missing User Warnings

Medium
Confidence
90% confidence
Finding
The guidance explicitly recommends collecting the full response including `errors`, exact documents, variables, and later suggests logging resolver parent objects. In GraphQL systems these artifacts often contain secrets, tokens, PII, authorization context, or tenant-specific data, so encouraging broad logging without redaction or scope limits can lead to sensitive-data exposure in logs and debugging tools.

Missing User Warnings

Medium
Confidence
95% confidence
Finding
Telling users to log the full parent object at a resolver can expose all fields loaded for that object, including private attributes not present in the GraphQL response. In a GraphQL debugging skill, this is more dangerous because resolver parents frequently carry backend-only data, auth-derived fields, and cross-tenant records, making accidental disclosure through logs plausible.

Missing User Warnings

Medium
Confidence
97% confidence
Finding
The skill explicitly instructs the agent to read from and write to persistent local files containing user preferences and prior context, but it provides no user-facing disclosure or consent step for that data access. This creates a privacy and transparency risk because sensitive project details, habits, stack information, or historical context may be silently persisted and later reused across sessions.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.