Security audit

UAPI 步骤2 (方法二): 获取原始文本接口

Security checks for vulnerabilities and agentic risk

Overview

This is a narrow API helper, but it can send a decryption key and plaintext clipboard contents through a third-party service with broad automatic triggers.

Install only if you intentionally want this agent to call UAPI's Clipzy raw-text endpoint. Avoid using it for secrets unless you are comfortable sending the decryption key to uapis.cn for server-side decryption, and invoke it explicitly with the endpoint or full Clipzy context rather than relying on ambiguous words like raw.

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

Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code

Findings (5)

Natural-Language Policy Violations

Medium

Confidence: 87% confidence
Finding: The natural-language description is primarily written in Chinese and frames the skill around Chinese operation naming, without stating that users may choose another language or locale. Under the policy, forcing a specific language without opt-in can be a natural-language policy violation.

Vague Triggers

Medium

Confidence: 96% confidence
Finding: The trigger list includes very generic terms such as "raw" and "clipzy" with no narrowing conditions, which can cause the orchestration layer to invoke this skill for unrelated requests. Because this skill is tied to retrieving raw/decrypted text from an API endpoint, over-triggering can route sensitive user requests into a data-access skill unexpectedly, increasing the risk of inappropriate data retrieval or disclosure.

Missing User Warnings

Medium

Confidence: 94% confidence
Finding: The skill explicitly instructs users to send their decryption key to the server so the server can return plaintext, but it provides no warning that this defeats end-to-end confidentiality and exposes key material to the service operator, logs, intermediaries, or monitoring systems. In the context of a clipboard/secrets retrieval workflow, this is a real privacy and security weakness because users may unknowingly disclose the very key intended to remain client-side.

Natural-Language Policy Violations

Low

Confidence: 84% confidence
Finding: All user-facing instructional text in this file is presented only in Chinese, with no indication of language choice or opt-in. Under the policy, forcing a specific language without offering user choice can be a natural-language policy violation unless the locale constraint is documented and justified.

Natural-Language Policy Violations

Low

Confidence: 80% confidence
Finding: The file presents the skill in Chinese while also prescribing English keywords and example utterances, but it does not state that language selection is optional or user-driven. Under the policy, forcing or assuming a language/locale without explicit opt-in can be a natural-language policy issue.

Static analysis

No suspicious patterns detected.