Back to skill

Security audit

Skills Auto Manager

Security checks for vulnerabilities and agentic risk

Overview

The skill is a disclosed skills manager, but it enables recurring automatic installation of other skills with weak approval and safety boundaries.

Review this skill before installing. Use it only if you are comfortable with a tool that can modify your installed skill set; preferably disable auto_install_low_risk and any cron job, require manual approval for every install/update, and review each candidate skill's publisher, permissions, version, and contents first.

Vulnerability Patterns
  • System PersistenceInstalls backdoors, hooks, services, or scheduled tasks that survive the run
  • 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
Findings (2)

T08 · Insecure Dependencies

Error
Location
implementation.md:74
Finding

Unsafe Autonomous Installation of Third-Party Skills

Content
View full analysis
= 4.5 && skill.downloads > 1000) return 'low'; if (riskFactors.length === 1) return 'medium'; return 'high'; } ``` ```javascript function decideInstall(skill, risk, config) { if (risk === 'low' && config.auto_install_low_risk) { return 'auto'; } if (risk === 'high' && config.auto_install_high_risk) { return 'auto'; } return 'confirm'; } ``` ```bash openclaw skills install ``` From `config.json`: ```json { "settings": { "auto_install_low_risk": true, "auto_install_high_risk": false, "backup_before_install": true } } ``` ### Technical Analysis The risk-assessment logic records dangerous properties such as API-key access, financial operations, beta status, and experimental status. However, it subsequently returns `low` if any of the following conditions are met: - The Skill is marked as official. - The Skill reports no dependencies. - The Skill has a rating of at least 4.5 and more than 1,000 downloads. These positive metadata conditions are evaluated before the accumulated risk factors. Therefore, a Skill that performs financial operations or requires API credentials can still be classified as low risk. Marketplace labels, download counts, ratings, and dependency counts do not establish that a Skill ...[truncated 2375 chars]
Remediation
View remediation

T06 · System Persistence

Warning
Location
implementation.md:170
Finding

Persistent Scheduled Execution Amplifies Supply-Chain Exposure

Content
View full analysis
Remediation
View remediation
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Rogue AgentSelf-Modification, Session Persistence
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Findings (20)

Self-Modification

High
Category
Rogue Agent
Confidence
85% confidence
Finding

Skill modifies its own code, configuration, or behavior at runtime. Self-modification enables an agent to escalate privileges, disable safety constraints, or install persistent backdoors.

Content

Scanner excerpt · implementation.md (reported line 211)May include surrounding context.

md
openclaw cron run skills-auto-manager-weekly

# 更新 cron job
openclaw cron update skills-auto-manager-weekly ...

# 删除 cron job
openclaw cron remove skills-auto-manager-weekly

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
96% confidence
Finding

This markdown file presents all operational instructions in Chinese, including headings, steps, troubleshooting, and checklist items, without indicating that the language is optional or region-specific. Under the language/locale policy, a skill document that effectively requires a specific language without opt-in is a natural-language policy violation.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
96% confidence
Finding

The file presents the skill description entirely in Chinese beginning with a Chinese-only headline, and nowhere offers an alternative language or opt-in choice. This can violate a language/locale policy when a skill effectively requires a specific language without user selection.

Content

No source excerpt is available for this finding.

File System Enumeration

Medium
Category
Data Exfiltration
Confidence
60% confidence
Finding

Code scans file system directories looking for sensitive files. This could be reconnaissance for credential theft.

Content

Scanner excerpt · README.md (reported line 224)May include surrounding context.

解决方案:

bash
# 检查权限
ls -la ~/.openclaw/skills/

# 查看 disk space
df -h

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The file’s user-facing instructions and descriptions are consistently presented only in Chinese, which effectively forces a specific language on users. There is no opt-in, alternative language option, or explanation that this skill is intentionally limited to a Chinese-speaking or region-specific audience.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
92% confidence
Finding

The description is entirely in Chinese, which imposes a specific language choice in user-facing metadata without indicating any language selection or opt-in. This can violate organizational language/locale policy when users are not offered a choice or told the skill is region-specific.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
84% confidence
Finding

The manifest advertises broad autonomous behavior such as checking updates, browsing the marketplace, recommending, and installing skills, but it does not define clear user approval gates, scope limits, or trigger constraints. In a skill-management context, vague automation increases the risk of unexpected package acquisition or execution paths, especially if later implementation assumes implicit consent.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
92% confidence
Finding

The manifest explicitly promotes auto-installation and cron-based automation while calling it 'safe' without documenting user-facing warnings, consent flow, rollback behavior, or supply-chain protections. For a skill that installs other skills, unattended scheduled execution materially increases the chance of silent installation of risky or compromised content from an external marketplace.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

This manifest sets the user language to "zh-CN" directly, which imposes a specific language/locale in configuration without indicating that the user can choose or opt in. The policy requires flagging language or locale constraints unless they are explicitly optional or clearly justified as region-specific.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The file states in Chinese that it is the implementation guide for AI execution, and the operational instructions throughout the document are written in Chinese. There is no indication that this locale is optional, user-selected, or required for a justified region-specific purpose.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The guide describes automatic installation of third-party skills and system changes without requiring a prominent user warning or explicit approval at the point of action. Installing code or agent capabilities automatically can introduce untrusted functionality, dependencies, or privilege use without informed consent.

Content

No source excerpt is available for this finding.

Context-Inappropriate Capability

Medium
Category
Not specified by scanner
Confidence
92% confidence
Finding

The implementation grants the skill authority to create, update, run, and delete persistent cron jobs, which extends beyond one-time checking/recommending and materially changes system behavior over time. Persistent automation increases attack surface because a compromised or misconfigured skill can repeatedly execute without fresh user approval.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
96% confidence
Finding

The scheduled cron-job setup enables persistent, repeated automated actions but does not clearly warn the user that it will continue running in the future. Hidden persistence is dangerous because it can repeatedly fetch, install, or modify skills long after the initial interaction, compounding the impact of mistakes or abuse.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

The manifest description is written only in Chinese ("自动管理 OpenClaw skills 的智能助手"), which indicates a language-specific presentation without any opt-in, alternative locale, or justification that the skill is region-specific. This can violate language/locale policy when users are not given a choice.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The file documents a weekly enabled cron job and repeated unattended execution but does not warn that background runs can perform actions without real-time user awareness. In this context, the danger is elevated because the scheduled task is tied to skill discovery and possible installation behavior, compounding persistence and supply-chain exposure.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

The document advertises '安全自动安装低风险 skills' without a clear warning that installation changes the local environment and may introduce code or automation the user did not manually review. In a skill whose purpose is to discover and install other skills, normalizing automatic installation materially increases supply-chain risk because users may infer the process is inherently safe from this summary text.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
91% confidence
Finding

This markdown file is natural-language content and appears to force a specific language/locale for all users. The policy requires flagging language or locale constraints when the skill does not offer user opt-in or explain that the Chinese-only format is intentionally region-specific.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
90% confidence
Finding

The entire skill report is written in Chinese, including headings, status descriptions, and action items, with no indication that the user opted into Chinese or that the skill is restricted to a Chinese-speaking or region-specific context. This can violate language/locale policy expectations for user-facing skill materials when no choice or justification is provided.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
91% confidence
Finding

The primary description is written entirely in Chinese, which can amount to a language/locale policy issue when no user opt-in or locale justification is provided. There is no indication that the skill supports language choice or that the locale restriction is intentional for a specific regional use case.

Content

No source excerpt is available for this finding.

Context-Inappropriate Capability

Low
Category
Not specified by scanner
Confidence
86% confidence
Finding

The manifest presents a general OpenClaw skills auto-manager, but the browsing phase narrows search terms to quantitative trading, stock, finance, and investment topics. That domain-specific targeting is not supported by the stated purpose and could bias recommendations toward a particular use case unrelated to generic skill management.

Content

No source excerpt is available for this finding.