T05 · Unauthorized Access and Privilege Escalation
Error
- Location
SKILL.md:287- Finding
Read-Only Safety Boundary Violated by Configuration-Changing Instructions
- Content
View full analysis
- Remediation
View remediation
Security audit
Network Topology Discovery
Security checks for vulnerabilities and agentic risk
Overview
The skill is mostly a documented network discovery workflow, but its read-only label conflicts with troubleshooting guidance that tells users to make network configuration changes.
Review before installing or using. Treat this as an authorized network-discovery playbook only, use read-only accounts, define a written discovery scope, protect generated reports as sensitive infrastructure data, and do not follow any configuration-changing recommendations without a separate approved change process.
Vulnerability Patterns
- Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs
- 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
Findings (2)
other
Warning
- Location
references/discovery-workflow.md:115- Finding
Recursive Collection of Security-Sensitive Network Topology Data
- Content
View full analysis
- Remediation
View remediation
Vulnerability Patterns
- MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
- 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
Findings (4)
Missing User Warnings
Medium
- Category
- Not specified by scanner
- Confidence
- 92% confidence
- Finding
The skill directs collection of hostnames, management IPs, software versions, routing data, ARP/MAC mappings, and adjacency relationships, but it does not warn that this creates a highly sensitive network inventory. Without explicit handling guidance, an agent or user may over-collect, persist, or expose topology and management-plane data that materially aids reconnaissance if leaked.
- Content
Intent-Code Divergence
Medium
- Category
- Not specified by scanner
- Confidence
- 95% confidence
- Finding
The skill is घोषित as read-only, but this section instructs operators to enable CDP/LLDP and modify VLAN or trunk configuration during troubleshooting. In a documentation-driven agent workflow, such change guidance can cause unauthorized configuration changes, expand attack surface, or violate change-control expectations because users may trust the skill's safety metadata and follow it operationally.
- Content
Intent-Code Divergence
Medium
- Category
- Not specified by scanner
- Confidence
- 96% confidence
- Finding
The troubleshooting guidance includes operational recommendations such as enabling LLDP, clearing tables, investigating connected devices, and enabling port security, which go beyond passive inspection. This contradicts the skill's read-only designation and can mislead an agent or operator into taking active network changes under the guise of a safe discovery workflow.
- Content
Missing User Warnings
Low
- Category
- Not specified by scanner
- Confidence
- 79% confidence
- Finding
This markdown file documents commands that reveal neighbor relationships, MAC/ARP tables, routing state, hostnames, serial numbers, management interfaces, and portions of running configuration. Although the file states the commands are read-only, it does not warn users that the collected output may contain sensitive infrastructure and inventory information that should be handled carefully.
- Content