Back to skill

Security audit

企业级智能客服知识库构建器

Security checks for vulnerabilities and agentic risk

Overview

This skill appears to be a local customer-service knowledge-base generator with disclosed inputs and no hidden network, file, credential, or persistence behavior.

Before installing, users should avoid submitting secrets, regulated data, or unredacted customer chat logs unless they are authorized to process that data in this environment. Human review is still needed before publishing the generated knowledge base, especially for finance, medical, legal, or customer-specific content.

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
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (6)

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The README explicitly encourages users to upload product documents, FAQs, and historical conversations, which commonly contain personal data, confidential business information, or regulated content. Because it provides no privacy notice, data-minimization guidance, redaction requirement, or handling constraints, users may submit sensitive material into the skill without understanding the exposure or compliance risk.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The skill explicitly accepts product documents, FAQs, and historical conversation data, which can easily include personal data, confidential business information, or regulated content. Because the skill description provides no privacy notice, data minimization guidance, retention limits, or handling constraints, users may upload sensitive material without understanding the exposure risk.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

This codebase presents the skill as an enterprise customer-service knowledge-base builder entirely in Chinese, and its templates, intents, fallback responses, and output text are all fixed to Chinese-language content. Because there is no user opt-in or configurable language/locale selection, the skill effectively forces a specific language, which matches the natural-language policy violation criteria.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

The manifest explicitly encourages uploading product documents, FAQs, and historical chat logs, which commonly contain personal data, confidential business information, or regulated content. There is no accompanying privacy notice, data minimization guidance, retention/deletion statement, or warning against submitting sensitive data, which increases the risk of accidental disclosure and non-compliant processing.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
84% confidence
Finding

All user-facing natural-language content in the file is Chinese, and there is no indication that the user can choose another language or that the locale restriction is intentional and documented. This may violate a language/locale policy requiring user choice or explicit justification.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
90% confidence
Finding

The natural-language description, triggers, workflow, and notes are all presented only in Chinese, which can amount to forcing a specific language without user opt-in. There is no indication that the skill is region-specific or that users may select another language.

Content

No source excerpt is available for this finding.