Google Tasks

• Outcome: Pass
• Latest audit: May 28, 2026, 7:38 AM
• Version:
  1.0.0

Overview

This is a coherent Google Tasks integration, but it can modify live tasks and stores OAuth tokens locally, so users should handle deletion and token files carefully.

Install only if you are comfortable granting read/write access to your Google Tasks. Keep credentials.json and token.json private, avoid shared or committed workspaces, and ask the agent to show the exact list and task before deleting anything.

SkillSpector

By NVIDIA

Vulnerability Patterns

• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• Trigger Abuse: Overly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands
• Privilege Escalation: Excessive Permissions, Sudo/Root Execution, Credential Access
• Supply Chain: Unpinned Dependencies, External Script Fetching, Obfuscated Code

Findings (6)

Missing User Warnings

Medium

• Confidence: 84% confidence
• Finding: The README explicitly documents task deletion but does not warn that the action is destructive or may be difficult to undo. In an agent skill context, this increases the chance of accidental data loss because users or higher-level agents may invoke deletion commands without understanding the consequences.

Vague Triggers

Medium

• Confidence: 83% confidence
• Finding: The description includes broad trigger phrases such as check, view, list, get, add, create, remove, and delete for general task-related requests. Overly broad invocation language can cause the skill to activate in situations the user did not specifically intend, which is more dangerous here because the skill supports destructive actions like deletion and authenticated API calls.

Missing User Warnings

Medium

• Confidence: 94% confidence
• Finding: The documentation explains how to delete tasks by number or title but does not warn that deletion is destructive or recommend an explicit confirmation step. In a skill that can act on live user data through OAuth, omission of confirmation guidance materially raises the chance of accidental or unauthorized task deletion.

Missing User Warnings

Medium

• Confidence: 87% confidence
• Finding: The setup documentation instructs users to persist OAuth credentials in `token.json` and notes that subsequent runs can use and refresh those credentials automatically, but it does not warn that this file contains bearer/refresh tokens that must be protected. If stored with weak filesystem permissions, committed to source control, or shared accidentally, an attacker could reuse the tokens to access and manipulate the user's Google Tasks without re-authenticating.

Missing User Warnings

Medium

• Confidence: 97% confidence
• Finding: The script performs an irreversible delete operation immediately once it resolves a task identifier, with no confirmation step, dry-run mode, or explicit safety interlock. In an agent skill context, this increases the chance of accidental or unintended deletion caused by ambiguous user input, wrong task indexing, or automation mistakes.

Missing User Warnings

Medium

• Confidence: 91% confidence
• Finding: The code writes OAuth tokens, including refresh tokens, to token.json on disk without warning the user or applying any file-permission hardening or secret storage protections. If the working directory is shared, backed up insecurely, or readable by other local users/processes, the stored token can be stolen and used to access or manipulate the user's Google Tasks data persistently.

VirusTotal

64/64 vendors flagged this skill as clean.