ClawHub
Back to skill

Security audit

China Cloud Deploy

Security checks for vulnerabilities and agentic risk

Overview

This China-cloud deployment guide is mostly purpose-aligned, but some default examples can expose cloud services or persistently change developer tooling without enough warning or scoping.

Review and adapt the commands before use. Treat the public endpoint and public bucket examples as intentionally public-only patterns, add authentication or gateway controls for APIs, use dedicated empty buckets for static assets, scan build output for secrets, and avoid persistent global registry changes unless you understand and can reverse them.

A.I.G

Vulnerability Patterns
  • Insecure DependenciesIntroduces malicious components through unsafe dependency sources
  • 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 (3)

T08 · Insecure Dependencies

Warning
Location
SKILL.md:18
Finding

Unpinned Package Installation and Persistent Third-Party Registry Configuration

Content
View full analysis
Remediation
View remediation

T09 · Insecure Skill Coding Practices

Error
Location
SKILL.md:59
Finding

Public Unauthenticated Serverless Endpoints Enabled by Default

Content
View full analysis
Remediation
View remediation

T09 · Insecure Skill Coding Practices

Error
Location
SKILL.md:120
Finding

Bucket-Wide Public Read Access Applied Without Content Validation

Content
View full analysis
Remediation
View remediation

SkillSpector

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 (3)

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
86% confidence
Finding

The instruction frames the skill as an expert specifically for deployment to Chinese cloud platforms and repeatedly centers China-specific operational assumptions, but it does not explicitly offer the user a language or locale choice. Under the policy, forcing a locale-specific mode without opt-in can be a natural-language policy violation unless clearly documented as a justified region-specific tool.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
92% confidence
Finding

The skill instructs users to set an OSS bucket ACL to public-read for static hosting, but it does not explicitly warn that all uploaded objects may become publicly accessible. In a deployment-focused skill, this omission can lead users to unintentionally expose sensitive build artifacts, backups, config files, or misuploaded secrets.

Content

No source excerpt is available for this finding.

External Transmission

Medium
Category
Data Exfiltration
Confidence
60% confidence
Finding

Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Content

Scanner excerpt · SKILL.md (reported line 263)May include surrounding context.

ICP Filing CLI Check

bash
# Check if domain has ICP filing
curl -s "https://hlwicpfwc.miit.gov.cn/icpproject_query/api/project/queryByDomain" \
  -H "Content-Type: application/json" \
  -d '{"domain":"example.com"}'