Back to skill

Security audit

continuous-test-review-loop

Security checks for vulnerabilities and agentic risk

Overview

This is a documentation-only methodology skill with some packaging and routing quality issues, but no evidence of hidden execution, data theft, persistence, or destructive behavior.

Install only if you want a Chinese-first methodology prompt for recurring test-review-improve cycles. Be aware that the broad triggers may activate it too often, and the licensing terms should be treated as unclear because the package says both MIT and all-rights-reserved for different parts.

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
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • 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
Findings (9)

Ae1

High
Category
analysis-evasion
Confidence
100% confidence
Finding

Referenced artifact was not completely inspected

Content

Scanner excerpt · SKILL.md (reported line 99)May include surrounding context.

md
| 文档与理论文本 | 本 `SKILL.md`、`README.md`、其中「理论依据」段与一切理论表述 | **不在 MIT 覆盖范围内**:保留所有权利(All rights reserved) |

Intent-Code Divergence

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

L009 将“.py 文件”表述为受 MIT 覆盖,L045 也写明 License: MIT;但 L098-L100、L116 又明确称本包无独立代码文件,且本包整体不适用任何代码许可、保留所有权利。这里不是单纯信息不全,而是同一技能文档对该技能资产许可范围给出了彼此冲突的意图说明。

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
96% confidence
Finding

The trigger keywords are extremely generic ('continuous', 'test', 'review', 'loop') and can match ordinary conversation unrelated to this skill. In an agent ecosystem, overly broad activation criteria can cause unintended skill invocation, prompt-routing errors, and context contamination, especially because this skill is methodology-oriented and could insert irrelevant process guidance into many tasks.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
98% confidence
Finding

The manifest-level trigger_keywords are too generic to safely distinguish this skill from normal discussion, increasing the chance of accidental activation across unrelated prompts. While the skill is not overtly malicious, broad triggers enlarge its routing surface and can interfere with correct tool/skill selection, which is a real integrity risk in multi-skill agent environments.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The manifest description embeds a long trigger list containing broad, generic phrases such as '改进', '打磨', and '验证方案', which are common in many unrelated conversations. This can cause unintended skill invocation or routing collisions, making the assistant apply this methodology in contexts where it was not explicitly requested, reducing reliability and potentially bypassing more appropriate skills.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
90% confidence
Finding

SQP-3 applies to all file types and covers language or locale policy violations. The explanatory licensing notice is presented in Chinese with only limited inline English glosses, and the file does not offer an alternative language version or indicate that the Chinese-only presentation is required for a specific regional context.

Content

No source excerpt is available for this finding.

Vague Triggers

Low
Category
Not specified by scanner
Confidence
78% confidence
Finding

This file is a markdown file, so SQP-1 applies. Line L07 defers key applicability details to another document section instead of making the boundaries self-contained here, which can create ambiguity about the exact scope and constraints of the licensing statement in this file.

Content

No source excerpt is available for this finding.

Intent-Code Divergence

Low
Category
Not specified by scanner
Confidence
77% confidence
Finding

L056-L064 主动宣称‘识别用户输入语言,自动用同语言回复’等运行时能力,但整份技能文件从 L098、L116 还说明这是‘纯文档/理论件’、无独立代码文件。该表述把未实现的能力写成既有功能,构成文档意图与实际内容的不一致。

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
88% confidence
Finding

The manifest description is entirely in Chinese and does not indicate that language selection is optional or that the skill is intentionally limited to Chinese-speaking users. This can violate language/locale policy when no user opt-in or justification is provided.

Content

No source excerpt is available for this finding.