ClawHub
Back to skill

Security audit

insta-orcha-task

Security checks across malware telemetry and agentic risk

Overview

This skill mostly matches its task-automation purpose, but it includes embedded API credentials and under-disclosed task/profile access that users should review before installing.

Review before installing. Do not use the embedded credentials; they should be removed and rotated. Run only with credentials you control, verify the task IDs and profile access are intended, inspect ZIP contents before upload, and avoid unattended cron execution until upload failures cannot be marked successful.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (1)

Tp4

High
Category
MCP Tool Poisoning
Confidence
94% confidence
Finding
The documented behavior claims the skill only grabs tasks, updates status, and uploads deliverables, but the finding indicates additional undisclosed capabilities: profile/tag enumeration, task filtering, manual task targeting, and helper behavior that may embed default credentials. Hidden task-selection logic and credential defaults materially expand the trust boundary, enabling unauthorized task access patterns, unintended data disclosure, or misuse of privileged API credentials.

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
api_client.py:88