Security audit

presales-cloud-architect

Security checks for vulnerabilities and agentic risk

Overview

This is a focused markdown-only sales architecture proposal skill with disclosed web research, no code execution, persistence, credential access, or hidden behavior.

Before using this skill, avoid pasting confidential customer architecture, nonpublic pain points, or sensitive commercial details unless you are comfortable with the agent using them to form web searches. For public-account planning and Chinese executive proposals, the behavior is coherent and low risk.

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

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 (2)

Missing User Warnings

Medium

Confidence: 95% confidence
Finding: The skill explicitly instructs the agent to use online search to enrich customer background based on user-provided company, website, architecture, and pain-point details, but it provides no warning, consent step, or minimization guidance before sending or deriving queries from that data. In a sales/pre-sales context, these inputs can contain confidential prospect information, making unintended disclosure to external search providers or connected tools a realistic privacy and confidentiality risk.

Natural-Language Policy Violations

Low

Confidence: 79% confidence
Finding: The entire instruction set is written to require a Chinese-language executive deliverable and a specific tone, with no indication that the user may choose another language or locale. Under the policy rule, forcing a specific language without opt-in can be a natural-language policy violation.

Static analysis

No suspicious patterns detected.