ClawHub
Back to skill

Security audit

OpenClaw TODO List

Security checks for vulnerabilities and agentic risk

Overview

This skill is a straightforward local TODO tracker that writes only its own TODO and DONE files, with some usability cautions around automatic updates and Chinese-only instructions.

Install this if you want a simple local TODO/DONE memory file workflow. Be aware it may prompt the agent to update task files proactively when work starts, progresses, or completes, and the instructions are Chinese-language oriented.

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
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • 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 (6)

Vague Triggers

Medium
Confidence
92% confidence
Finding
The activation conditions are broad enough that the skill could trigger during ordinary task-oriented conversation, causing unsolicited reads or writes to the todo memory files. In a memory-managing skill, over-triggering can lead to incorrect persistence of user content, unintended state changes, and confusion about what the agent is doing.

Natural-Language Policy Violations

Medium
Confidence
88% confidence
Finding
The skill metadata and content are written entirely in Chinese and do not indicate language negotiation or user preference handling, which can pressure responses or record formats into a specific language without opt-in. This is primarily a usability and policy issue, but it can indirectly cause incorrect task recording if the user operates in another language.

Skill Enumeration

Medium
Category
Agent Snooping
Content
|------|------|
| 待办 | `memory/todo/TODO.md` |
| 已完成 | `memory/todo/DONE.md` |
| 技能说明 | `skills/todo-list/SKILL.md` |
| 详细规范 | `skills/todo-list/references/spec.md` |
| 本示例 | `skills/todo-list/references/example-todo.md` |
Confidence
80% confidence
Finding
Skill enumerates or reads other installed skills. Access to other skills' SKILL.md files or the skills directory reveals prompt instructions, capabilities, and secrets that should be invisible to peer skills.

Vague Triggers

Low
Confidence
80% confidence
Finding
The example trigger '有新的 TODO 需要记录吗' is vague and can overlap with normal conversational planning, increasing the chance that the skill activates without clear user consent. Although the direct impact is limited to todo tracking, this still risks unintended memory writes and inaccurate task state.

Natural-Language Policy Violations

Low
Confidence
93% confidence
Finding
The natural-language instructions, examples, and headings are entirely in Chinese, which can impose a language expectation on users. There is no statement that the skill is Chinese-only, region-specific, or that users may opt into another language.

Missing User Warnings

Low
Confidence
87% confidence
Finding
This markdown file instructs the skill to create directories and files and later update TODO.md and DONE.md, which affects user data on disk. The description does not include any user-facing warning or disclosure that local files under memory/todo/ will be created and modified.

Static analysis

No suspicious patterns detected.