ClawHub
Back to skill

Security audit

Bebebebebe

Security checks across malware telemetry and agentic risk

Overview

This setup guide is not malicious, but it can give an agent broad paid API access and live connected-account actions without clear confirmation safeguards.

Install only if you want Orthogonal to become a broad API and integration layer for your agent. Protect the API key, prefer environment or secure secret storage when possible, review any added skills separately, connect only the OAuth services you need, and require manual approval before any paid API call, email, Slack post, calendar change, repository action, or Drive/Sheets/Notion update.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Rogue AgentSelf-Modification, Session Persistence
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (4)

Missing User Warnings

Medium
Confidence
89% confidence
Finding
The skill documents storing an API key in a local JSON file and provides a manual plaintext write example without warning about local secret exposure risks. While this is common CLI behavior, omitting guidance on file permissions, secret handling, and safer alternatives can lead users or downstream agents to persist reusable credentials insecurely on shared systems.

Missing User Warnings

Medium
Confidence
86% confidence
Finding
The integration section gives ready-to-run examples for sending email, creating calendar events, and posting Slack messages through connected third-party accounts, but does not clearly warn that these commands perform real external actions and may disclose data or modify user resources. In an agent setting, this increases the chance of unintended side effects if an agent copies examples or invokes actions without explicit user confirmation.

Credential Access

High
Category
Privilege Escalation
Content
## Authentication

Your API key is stored at `~/.config/orthogonal/credentials.json` after running `orth login`.

**Manual setup** (if needed):
Confidence
92% confidence
Finding
The skill explicitly identifies the credential storage path `~/.config/orthogonal/credentials.json`, which tells an agent or user exactly where a reusable API secret resides. In a skill ecosystem, exposing the location of sensitive material meaningfully lowers the barrier to credential discovery, misuse, or exfiltration by other tools, prompts, or compromised components.

Session Persistence

Medium
Category
Rogue Agent
Content
**Manual setup** (if needed):

1. Get your API key at https://orthogonal.com/dashboard/settings
2. Create the credentials file:

```bash
mkdir -p ~/.config/orthogonal
Confidence
90% confidence
Finding
The manual setup instructs writing a long-lived API key directly to a persistent local file, creating session persistence that survives shell sessions and may remain available indefinitely. If the host is shared, backed up insecurely, or later compromised, the retained credential can be reused to access paid APIs and connected services.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.