Back to skill

Security audit

使用 LLM 动态构建和维护知识库

Security checks for vulnerabilities and agentic risk

Overview

This skill is a disclosed Markdown knowledge-base assistant that reads and updates a user-designated Knowledge wiki, with no evidence of hidden execution, exfiltration, or persistence.

Install this only for a Knowledge-style Markdown wiki you want an agent to read and update. Grant access to the intended Knowledge directory, review bulk rename or restructuring plans before approval, and keep source materials read-only unless you explicitly want them changed.

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
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • 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 (3)

Ae1

High
Category
analysis-evasion
Confidence
100% confidence
Finding

Referenced artifact was not completely inspected

Content

Scanner excerpt · SKILL.md (reported line 303)May include surrounding context.

md
发布 skill 时,应将 `Knowledge/schema/scripts/verify_wiki.py` 作为 skill 包内的 `scripts/verify_wiki.py` 随 `SKILL.md` 一起打包;根目录 wrapper 只属于当前 vault 的便利入口。

Undeclared Tool Scope

Medium
Category
MCP Least Privilege
Confidence
83% confidence
Finding

The skill describes operational behavior that requires reading local files (for example scanning source/, reading wiki/index.md, and inspecting many Markdown pages) but does not declare an explicit tool scope such as permissions or allowed-tools. That mismatch can cause the runtime or calling agent to infer broader access than users expect, reducing transparency and weakening least-privilege controls around local knowledge-base contents.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The document states '默认目录使用中文' and further requires directory and file names to use Chinese wherever possible, while discouraging English directory names. This is a natural-language locale policy constraint that forces a specific language preference rather than offering the user a choice or framing it as an optional convention.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.