ClawHub
Back to skill

Security audit

Twenty CRM OAuth Mastery

Security checks across malware telemetry and agentic risk

Overview

This documentation-only OAuth troubleshooting skill is coherent, but it recommends risky production authentication changes that could expose user sessions if followed as written.

Use this only as a high-sensitivity OAuth troubleshooting reference, not as drop-in production guidance. Before applying its fixes, review the httpOnly false cookie recommendation with a security engineer, prefer HttpOnly/server-managed sessions where possible, and never paste live OAuth secrets, token values, or raw docker env output into chats, tickets, logs, or documentation.

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
95% confidence
Finding
The README explicitly recommends setting `httpOnly: false` as a quick fix for OAuth cookie handling, but provides no warning that this makes the token-bearing cookie readable by client-side JavaScript. In an OAuth context, that materially increases exposure to token theft via XSS, third-party scripts, browser extensions, or other client-side compromise, and the skill context makes this more dangerous because it presents the change as a proven solution for production troubleshooting.

Missing User Warnings

High
Confidence
99% confidence
Finding
The skill explicitly instructs setting an OAuth token cookie with `httpOnly: false` so frontend JavaScript can read it. Storing bearer tokens in a script-readable cookie materially increases exposure to token theft through XSS, browser extensions, or other injected client-side code, and the skill presents this as a required fix without warning about the security tradeoff.

Missing User Warnings

Medium
Confidence
95% confidence
Finding
The skill includes a real-looking OAuth client ID and especially a client secret example in configuration guidance without any warning that secrets must never be committed, shared in docs, or reused. This can normalize unsafe credential handling and, if the values are real, directly leak sensitive credentials that could be abused to impersonate the application in OAuth flows.

VirusTotal

64/64 vendors flagged this skill as clean.

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:92