Back to skill

Security audit

Google Tasks

Security checks for vulnerabilities and agentic risk

Overview

This is a straightforward Google Tasks skill, with expected account-linking and task mutation features but some install and deletion caution needed.

Install only if you trust the MorphixAI plugin and understand it will use a MorphixAI API key with a linked Google Tasks account. Treat create, update, complete, and delete examples as live changes to your Google Tasks data, and use extra care with delete actions.

Vulnerability Patterns
  • Insecure DependenciesIntroduces malicious components through unsafe dependency sources
  • 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
Findings (1)

T08 · Insecure Dependencies

Warning
Location
SKILL.md:17
Finding
Unpinned Third-Party Plugin Installation## Vulnerability Details **File Location**: `SKILL.md`, line 17 **Vulnerability Type**: Unpinned and unauditable third-party dependency **Risk Level**: Medium **Code Snippet**: ```text 1. **安装插件**: `openclaw plugins install openclaw-morphixai` ``` ### Technical Analysis The documented installation command does not specify an immutable plugin version or integrity checksum. The project also contains no source code, lockfile, package manifest, or reproducible source reference for `openclaw-morphixai`. Therefore, the exact code installed by this command may change after the Skill has been reviewed. The plugin is expected to operate in an environment containing `MORPHIXAI_API_KEY` and to mediate access to a linked Google Tasks account. A compromised publisher account, malicious future release, registry compromise, or dependency substitution could consequently introduce behavior outside the documented Google Tasks integration. The reviewed file transparently identifies MorphixAI as the intermediary, and no direct evidence of a malicious package, hidden endpoint, credential theft, or unauthorized execution was found. The risk arises from installing mutable third-party code whose implementation is outside the audit scope. ### Attack Path 1. A user follows the instruction at `SKILL.md:17`. 2. The package manager resolves the current release of `openclaw-morphixai` because no reviewed version or digest is specified. 3. An attacker compromises the package publisher, registry, distribution channel, or a future package release. 4. The affected release is installed and executes with the permissions granted to OpenClaw plugins. 5. The malicious plugin may access the configured MorphixAI credential or abuse the linked Google Tasks operations available through the integration. ### Impact Assessment Successful exploitation could expose the MorphixAI API key and task data available to the linked account. It could also allow unauthorized creation, modification, completio ...[truncated 240 chars]
Remediation
## Remediation Suggestions 1. Pin `openclaw-morphixai` to a specific, reviewed, immutable version rather than installing the latest available release. 2. Verify the package with a trusted cryptographic digest or signature and document the expected publisher identity. 3. Include a lockfile, package manifest, or reproducible source reference that identifies the exact audited implementation. 4. Review and document the plugin's required permissions, network destinations, credential access, and Google Tasks authorization scopes. 5. Restrict the plugin to the minimum privileges required for Google Tasks operations. 6. Store `MORPHIXAI_API_KEY` in an approved secret store and prevent unnecessary processes or plugins from reading it. 7. Establish an update-review process so newer plugin releases are audited before deployment.
Vulnerability Patterns
  • 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
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Findings (2)

Missing User Warnings

Medium
Confidence
94% confidence
Finding
The skill documents destructive `delete_task_list` and `delete_task` operations without warning that they permanently remove remote Google Tasks data. In an agent setting, this increases the chance of accidental or socially engineered data loss because users may not realize the action is irreversible or state-changing.

Missing User Warnings

Low
Confidence
88% confidence
Finding
The workflow and notes describe update and completion actions without clearly warning that they modify the user's live Google Tasks account. This can lead to unintended state changes if an agent executes examples directly or if users interpret the skill as read-only documentation.

Static analysis

No suspicious patterns detected.