ClawHub
Back to skill

Security audit

Skill Manage

Security checks for vulnerabilities and agentic risk

Overview

This is a real OpenClaw skill manager, but its uninstall cleanup can remove files beyond the selected skill and can delete config files even when the prompt says they are preserved.

Review this before installing. Use --dry-run first, avoid --force, back up skill config files, and verify every path the uninstall command plans to remove. The cleanup logic should be narrowed to recorded skill-owned paths before it is trusted for routine use.

A.I.G

Vulnerability Patterns
  • Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs
  • Insecure Skill Coding PracticesFinds exploitable flaws such as hardcoded secrets or command injection
  • 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)

T09 · Insecure Skill Coding Practices

Warning
Location
scripts/skill_manage.py:246
Finding

Heuristic Residual Cleanup Can Delete Unrelated Files Outside the Selected Skill

Content
View full analysis
Remediation
View remediation

T05 · Unauthorized Access and Privilege Escalation

Note
Location
scripts/skill_manage.py:226
Finding

Uninstall Reconnaissance Reads Complete Global Configuration Files

Content
View full analysis
Remediation
View remediation

SkillSpector

Vulnerability Patterns
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Behavioral ASTexec() Call, eval() Call, Dynamic Import
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Findings (8)

Credential Access

High
Category
Privilege Escalation
Confidence
60% confidence
Finding

Code accesses credential files (SSH keys, AWS credentials, etc.). This could indicate credential theft attempts.

Content

Scanner excerpt · README.md (reported line 40)May include surrounding context.

md
`uninstall` 命令在删除主目录前会先**扫描残留痕迹**，包括：

- ✅ SkillHub 锁文件记录（从 `skillhub.lock.json` 移除）
- ✅ 配置文件（config.json / .env / .yaml）
- ✅ 缓存目录（cache/、TMP/、__pycache__/）
- ✅ workspace 中的残留文件
- ✅ OpenClaw 主配置引用

Credential Access

High
Category
Privilege Escalation
Confidence
60% confidence
Finding

Code accesses credential files (SSH keys, AWS credentials, etc.). This could indicate credential theft attempts.

Content

Scanner excerpt · scripts/skill_manage.py (reported line 243)May include surrounding context.

python
`uninstall` 命令在删除主目录前会先**扫描残留痕迹**，包括：

- ✅ SkillHub 锁文件记录（从 `skillhub.lock.json` 移除）
- ✅ 配置文件（config.json / .env / .yaml）
- ✅ 缓存目录（cache/、TMP/、__pycache__/）
- ✅ workspace 中的残留文件
- ✅ OpenClaw 主配置引用

Credential Access

High
Category
Privilege Escalation
Confidence
60% confidence
Finding

Code accesses credential files (SSH keys, AWS credentials, etc.). This could indicate credential theft attempts.

Content

Scanner excerpt · scripts/skill_manage.py (reported line 244)May include surrounding context.

python
residuals["has_credentials"] = True
        residuals["config_files"].append(str(config_file))
    # 扫描主目录下的所有 .json / .env / .yaml 文件
    for pattern in ["*.json", "*.env", "*.yaml", "*.yml", "*.ini", "*.cfg"]:
        for f in skill_path.rglob(pattern):
            if f.is_file():
                residuals["config_files"].append(str(f))

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

The trigger phrase "帮我管理 Skill" is broad, conversational language that can overlap with normal help-seeking requests rather than a narrowly scoped command. In a system-management skill that can update or uninstall installed skills, overly broad activation increases the chance of accidental invocation and unintended administrative actions.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

This code file contains multiple natural-language strings that force Chinese for key user interactions, including the module description and later command prompts/help. Under the stated policy, forcing a specific language without user opt-in is a natural-language policy violation unless the locale restriction is clearly justified.

Content

No source excerpt is available for this finding.

subprocess module call

Medium
Category
Dangerous Code Execution
Confidence
70% confidence
Finding

subprocess module calls execute external commands. Without careful input validation, this enables command injection.

Content

Scanner excerpt · scripts/skill_manage.py (reported line 183)May include surrounding context.

python
if meta["source"] == "GitHub":
        print(f"[GitHub] {meta['name']} -> git pull")
        try:
            r = subprocess.run(
                ["git", "pull"],
                cwd=str(skill_path),
                capture_output=True,

subprocess module call

Medium
Category
Dangerous Code Execution
Confidence
70% confidence
Finding

subprocess module calls execute external commands. Without careful input validation, this enables command injection.

Content

Scanner excerpt · scripts/skill_manage.py (reported line 199)May include surrounding context.

python
slug_arg = meta["name"].lower().replace(" ", "-")
        print(f"[SkillHub] {meta['name']} -> skillhub install {slug_arg}")
        try:
            r = subprocess.run(
                ["skillhub", "install", slug_arg],
                capture_output=True,
                text=True,

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
90% confidence
Finding

This markdown file includes section headings and entries in Chinese while earlier content is in English, but it does not explain whether the skill is intentionally bilingual or scoped to Chinese-speaking users. Under the language/locale policy rule, forcing or assuming a language without opt-in can be a policy concern.

Content

No source excerpt is available for this finding.