Back to skill

Security audit

Ai Fact Checker

Security checks for vulnerabilities and agentic risk

Overview

This fact-checking skill is mostly purpose-aligned, but its script can turn text being checked into shell command execution.

Review before installing. The skill’s goal is coherent, but it should replace execSync string interpolation with execFileSync or another argument-array API, add input length and timeout controls, and warn users not to submit secrets or confidential text to online search.

Vulnerability Patterns
  • Insecure DependenciesIntroduces malicious components through unsafe dependency sources
  • Insecure Skill Coding PracticesFinds exploitable flaws such as hardcoded secrets or command injection
  • 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
Findings (2)

T09 · Insecure Skill Coding Practices

Error
Location
scripts/fact-check.js:28
Finding

Shell Command Injection Through Fact-Check Input

Content
View full analysis
Remediation
View remediation

T08 · Insecure Dependencies

Warning
Location
SKILL.md:98
Finding

Unpinned Package Execution in Installation Instructions

Content
View full analysis
Remediation
View remediation
install ai-fact-checker ``` - Replace `` with an exact version rather than a range or tag such as `latest`. - Document the expected package registry and avoid untrusted registry overrides. - Prefer a previously installed and independently verified CLI in controlled environments. - Where supported, verify package signatures, checksums, provenance attestations, or registry integrity metadata before execution. - Maintain a documented update process that reviews new installer versions before changing the pinned version. ]]>
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • YARA SignaturesMalware Match, Webshell Match, Cryptominer Match
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Findings (12)

YARA rule 'agent_skill_mcp_tool_poisoning_metadata': MCP/tool metadata poisoning indicators in tool schemas or skill manifests [agent_skills]

High
Category
YARA Match
Confidence
80% confidence
Finding

YARA rule matched a hack tool or exploit indicator (offensive tools, reconnaissance, privilege escalation, or exploit frameworks).

Content

Scanner excerpt · SKILL.md (reported line 3)May include surrounding context.

md
---
name: ai-fact-checker
description: 🔍 AI 事实核查 - 自动验证大模型输出准确性,检测 hallucination 幻觉编造内容,联网搜索交叉验证给出可信度评分和修正。解决大模型一本正经胡说八道的痛点。
---

# 🔍 AI Fact Checker - AI 事实核查

自动验证 AI 输出信息的真实性,检测幻觉(hallucination)编造内容,联网搜索交叉验证,给出可信度评分,并自动修正错误信息。

## ✨ 痛点解决

大模型经常一本正经地**胡说八道**:编造不存在的论文、数据、人名、事件、网址。这个技能帮你:

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The skill explicitly says it performs online cross-verification via search, but the user-facing description does not clearly warn that submitted content will be sent to external search services. Users may paste sensitive, proprietary, or regulated data into the fact-checker without realizing it will leave the local environment.

Content

No source excerpt is available for this finding.

Rp1

Medium
Category
MCP Rug Pull
Confidence
92% confidence
Finding

The installation command uses npx clawhub without pinning a specific version, so users may execute whatever package version is current at install time. That creates a supply-chain risk: a compromised or maliciously updated package could run arbitrary code on the user's machine during install or execution.

Content

No source excerpt is available for this finding.

Context-Inappropriate Capability

Medium
Category
Not specified by scanner
Confidence
91% confidence
Finding

The skill builds a shell command with user-derived content and executes it via execSync, which invokes a shell. Although double quotes are escaped, shell metacharacters such as backticks or command substitution syntax are not neutralized, so crafted input can break the intended boundary and trigger arbitrary local command execution. In a fact-checking skill, this is especially dangerous because it processes untrusted free-form text by design.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The skill sends user-provided text to an external search service without explicit notice or consent. Because the input can contain sensitive or proprietary content, this creates a privacy and data-exfiltration risk, and a fact-checking workflow makes such transmission likely and repeated for multiple extracted statements.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
91% confidence
Finding

The report verdicts, reasons, and summary strings are all emitted in Chinese, forcing a specific language regardless of user preference. This is a natural-language policy concern unless the locale restriction is explicitly documented and justified, which is not present in this file.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
81% confidence
Finding

The skill's description and operating instructions are presented entirely in Chinese, while the file does not indicate that the skill is region-specific or that users may choose another language. This can violate a language/locale policy when a skill implicitly forces one language without opt-in.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
84% confidence
Finding

This text is written entirely in Chinese and does not indicate that the skill is region-specific or that users can choose another language. Under the language/locale policy rule, forcing a specific language without opt-in can be a policy concern.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
95% confidence
Finding

The package description is written entirely in Chinese, which indicates a language-specific presentation without any visible opt-in or alternative language option. Under the policy, forcing a specific language without user choice can be a natural-language locale violation unless the regional constraint is explicitly documented and justified.

Content

No source excerpt is available for this finding.

Intent-Code Divergence

Low
Category
Not specified by scanner
Confidence
76% confidence
Finding

The comment frames this as use of a web search tool, which implies a direct tool integration, but the code actually constructs a shell command and executes it with execSync. That documentation omits and thereby misrepresents the significant behavior of invoking a subprocess on the host.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
95% confidence
Finding

This JavaScript test file includes user-facing natural-language content in Chinese, and subsequent console output also uses Chinese phrases for statements under test. Under the policy, forcing a specific language without user opt-in can be a natural-language policy violation, and there is no indication here that the skill offers locale selection or that the language restriction is justified.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
91% confidence
Finding

The test emits multiple Chinese-language statements directly in console output, reinforcing a fixed-language experience. Because the file provides no mechanism for user language selection and no documented justification for the locale constraint, this fits the language-policy concern described by SQP-3.

Content

No source excerpt is available for this finding.