ClawHub
Back to skill

Security audit

Andrew Google Tasks

Security checks across malware telemetry and agentic risk

Overview

This is a disclosed Google Tasks helper that can read and change tasks after OAuth authorization, with local credential files users should protect.

Install only if you are comfortable granting this skill access to read, create, modify, complete, and delete Google Tasks for the selected account. Keep ~/.google-credentials.json and ~/.google-tasks-token.pickle private, do not use token pickle files from untrusted sources, and revoke the OAuth grant or delete the token file when you no longer use the skill.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Findings (3)

Credential Access

High
Category
Privilege Escalation
Content
## Security Notes

- OAuth ����� `~/.google-tasks-token.pickle` � �����
- ������� ��� `~/.google-credentials.json` � ����� (�린�, ��� ���과 공�)
- � ���� `.gitignore` � ����� ����
- �� ��: `https://www.googleapis.com/auth/tasks` (Tasks �체 �근)
Confidence
86% confidence
Finding
The skill documents the location of sensitive OAuth artifacts, including a token pickle and shared credentials file in the user's home directory. Exposing credential file paths and encouraging reuse across skills increases the chance that other tools, skills, or users will access or mishandle these secrets, enabling unauthorized API access.

Credential Access

High
Category
Privilege Escalation
Content
## Troubleshooting

**"OAuth ������� �� ��� ����" ��:**
- `~/.google-credentials.json` ��� ��� ��
- 구� �린� ��� �� � �미 ��� �� �����

**�� ���:**
Confidence
84% confidence
Finding
The troubleshooting section again references the sensitive OAuth credentials file in a way that normalizes direct user interaction with secret material. Repeatedly surfacing credential locations increases the likelihood of accidental disclosure, unsafe copying, or use by unrelated tooling, especially in an agent environment where file paths can become actionable targets.

Credential Access

High
Category
Privilege Escalation
Content
## Integration with Other Google Skills

Same OAuth credentials (`~/.google-credentials.json`) are shared with `google-calendar` and `google-sheets` skills, so you only need to authenticate once!
Confidence
96% confidence
Finding
The statement that OAuth credentials are shared with other Google skills materially increases risk by promoting cross-skill credential reuse. In an agent ecosystem, this breaks isolation boundaries: compromise or misuse in one skill can grant access to other Google services authenticated with the same credential set.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.