ClawHub
Back to skill

Security audit

Unitask Agent

Security checks for vulnerabilities and agentic risk

Overview

This is a disclosed Unitask MCP connector that uses a scoped API token to read and modify the user's own tasks, with no hidden code or persistence mechanisms found.

Install only if you are comfortable giving an agent the selected Unitask scopes. Prefer read-only at first, add write/delete only when needed, keep the API token in a secret store, and confirm before applying deletes, merges, moves, or schedule changes.

A.I.G

Vulnerability Patterns
  • Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
  • Agent Memory PoisoningWrites attacker-controlled rules into memory that affect later sessions
  • Remote Payload Retrieval and ExecutionFetches external code whose behavior can change after review
  • Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally
  • Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs

SkillSpector

Vulnerability Patterns
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Rogue AgentSelf-Modification, Session Persistence
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (3)

Session Persistence

Medium
Category
Rogue Agent
Content
Start finishing tasks instead of just organizing them: connect any AI agent to Unitask (unitask.app) to manage and do your tasks with secure prioritization, tags, time blocks and more.

Unitask is now in **public beta**. Anyone can sign up at `https://unitask.app` and create a scoped API token.

## What it does
Confidence
60% confidence
Finding
Skill establishes unauthorized persistence across sessions via cron jobs, startup scripts, or state files. Session persistence allows an attacker to maintain access beyond the current interaction.

Autonomous Decision Making

Medium
Category
Excessive Agency
Content
2. User creates a Unitask API token from `Unitask -> Dashboard -> Settings -> API`.
3. User stores it in their agent/app secret store as: `UNITASK_API_KEY=<token>`.

Never ask users to paste full tokens in chat logs.

## Scope model
Confidence
80% confidence
Finding
Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.

Autonomous Decision Making

Medium
Category
Excessive Agency
Content
2. User creates a Unitask API token from `Unitask -> Dashboard -> Settings -> API`.
3. User stores it in their agent/app secret store as: `UNITASK_API_KEY=<token>`.

Never ask users to paste full tokens in chat logs.

## Scope model
Confidence
80% confidence
Finding
Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.

Static analysis

No suspicious patterns detected.