Back to skill

Security audit

AI-CLI Design

Security checks for vulnerabilities and agentic risk

Overview

This is a documentation-only skill about designing AI-friendly CLI tools, with no hidden execution, persistence, credential use, or destructive behavior in the artifact.

Before using this skill as guidance, review destructive CLI examples carefully and add explicit user safeguards in any real tool you build. Non-Chinese users may also want translated or localized documentation, since most of the body is Chinese-first.

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
  • Tool MisuseTool Parameter Abuse, Chaining Abuse, Unsafe Defaults
  • 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 (2)

Tool Parameter Abuse

High
Category
Tool Misuse
Confidence
80% confidence
Finding

Tool parameters are crafted to achieve unintended or unsafe behavior. Parameter abuse can bypass intended safety checks (e.g. shell=True, --force, dangerous glob patterns).

Content

Scanner excerpt · SKILL.md (reported line 148)May include surrounding context.

md
| `tool list --json` | `GET /items` | 查询 |
| `tool get ID --json` | `GET /items/:id` | 获取 |
| `tool create --name foo --json` | `POST /items` | 创建 |
| `tool delete ID --json` | `DELETE /items/:id` | 删除 |
| `tool scan --json` | `POST /scan` | 动作 |

**设计原则:**

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
91% confidence
Finding

The skill body is written predominantly in Chinese and presents the guidance as a bilingual title with Chinese-first instructional content, but it does not state that language choice is optional or user-selectable. This can violate a language/locale policy when a skill implicitly constrains outputs or usage to a specific language without opt-in.

Content

No source excerpt is available for this finding.