Back to skill

Security audit

ai-reply-heuristics

Security checks for vulnerabilities and agentic risk

Overview

This appears to be a local heuristic text scorer, but its declared tool access is broader than its purpose and conflicts with its own no-network/no-filesystem/no-subprocess permission claims.

Review the requested tool permissions before installing. The actual code is a simple local scorer, but the skill should narrow or explain its allowed tools, especially Write and WebFetch, and should clarify language support and file-vs-text usage.

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 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 (7)

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
84% confidence
Finding

The file title and all operational guidance are presented in Chinese, and there is no indication that language selection is optional or limited to a region-specific use case. Under the policy, forcing a specific language without user opt-in is a natural-language policy concern.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
90% confidence
Finding

The primary description field is written in Chinese, while the skill targets multiple platforms and does not state that it is China-specific or requires Chinese output. Under the policy, forcing a specific language without user opt-in is a natural-language policy violation unless the locale constraint is explicitly justified.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

This code includes natural-language instructions, disclaimers, argument help text, and result labels entirely in Chinese. Under the policy, forcing a specific language without user opt-in is a locale-policy issue unless the restriction is explicitly documented and justified, which is not present here.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
79% confidence
Finding

The file presents core skill information in both Chinese and English, with several important notices primarily in Chinese, but it does not state whether the skill is intended for Chinese-speaking users only or offer an explicit language preference/opt-in. Under the policy rule for language or locale constraints, this can be considered a mild natural-language policy issue because users are not given a clear choice.

Content

No source excerpt is available for this finding.

Intent-Code Divergence

Low
Category
Not specified by scanner
Confidence
83% confidence
Finding

The English usage text says to 'Pass the text as an argument', which describes a single direct-text CLI pattern. Elsewhere in the same file, the interface and input schema declare two distinct inputs, file and text, so the documentation does not match the declared behavior contract and may mislead users about how the skill is actually invoked.

Content

No source excerpt is available for this finding.

Intent-Code Divergence

Low
Category
Not specified by scanner
Confidence
88% confidence
Finding

Line L090 states that the caller must provide the text string as a parameter, implying text-only operation. However, the declared input schema at L036-L044 includes both file and text, so the documentation narrows and misstates the intended behavior.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
75% confidence
Finding

The phrase '签署免责确认' introduces a procedural requirement in natural language that may impose a locale- or organization-specific process without explaining applicability. While not a direct security issue, it reads as a policy-like requirement that is not scoped or justified for all users of the skill.

Content

No source excerpt is available for this finding.