ClawHub
Back to skill

Security audit

Chinese LLM Models (Kimi 2.5, MiniMax 2.5, Qwen, DeepSeek) with One Key

Security checks for vulnerabilities and agentic risk

Overview

The skill is a coherent AIsa setup guide, but it should be reviewed because it encourages risky API-key handling and overstates privacy for an external model provider.

Install only if you intend to route OpenClaw model prompts and outputs through AIsa. Treat AISA_API_KEY as a secret: prefer the interactive onboarding flow or a protected secret store, avoid passing the key in command-line arguments, and do not run echo $AISA_API_KEY in shared terminals or logs. Verify AIsa and Moonshot privacy terms yourself before sending regulated or highly sensitive data.

A.I.G

Vulnerability Patterns
  • 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
  • Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally
Findings (1)

T09 · Insecure Skill Coding Practices

Warning
Location
SKILL.md:38
Finding

API Credential Exposure Through Command-Line Arguments, Shell History, and Terminal Output

Content
View full analysis
Remediation
View remediation

SkillSpector

Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • 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
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (19)

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The skill description includes broad activation triggers such as comparing pricing, switching models, troubleshooting provider configuration, and reacting to generic mentions of API keys or model families. This can cause the skill to activate in unrelated conversations and steer users toward this provider, increasing the chance of inappropriate external-provider guidance or unintended data routing.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
92% confidence
Finding

The guide tells users to pass the API key via an environment variable and directly on the command line, including --aisa-api-key "你的密钥", but gives no warning that credentials are sensitive. Command-line arguments can be exposed through shell history, process listings, logs, or terminal recordings, which increases the chance of accidental secret leakage.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The statement that users '完全不用担心数据隐私问题' is an absolute privacy assurance that is not technically safe to make. Even if a zero-data-retention agreement exists, risks can remain in transit, logging layers, customer misconfiguration, legal changes, or upstream provider behavior, so unconditional claims may cause users to send sensitive data under false assumptions.

Content

No source excerpt is available for this finding.

External Transmission

Medium
Category
Data Exfiltration
Confidence
50% confidence
Finding

Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Content

Scanner excerpt · AIsa Chinese Models v3/SKILL.md (reported line 51)May include surrounding context.

md
"models": {
    "providers": {
      "aisa": {
        "baseUrl": "https://api.aisa.one/v1",
        "apiKey": "${AISA_API_KEY}",
        "api": "openai-completions",
        "models": [

External Transmission

Medium
Category
Data Exfiltration
Confidence
50% confidence
Finding

Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Content

Scanner excerpt · AIsa Chinese Models v3/SKILL.md (reported line 219)May include surrounding context.

md
"models": {
    "providers": {
      "aisa": {
        "baseUrl": "https://api.aisa.one/v1",
        "apiKey": "${AISA_API_KEY}",
        "api": "openai-completions",
        "models": [

External Transmission

Medium
Category
Data Exfiltration
Confidence
50% confidence
Finding

Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Content

Scanner excerpt · AIsa Chinese Models v3/SKILL.md (reported line 238)May include surrounding context.

md
"models": {
    "providers": {
      "aisa": {
        "baseUrl": "https://api.aisa.one/v1",
        "apiKey": "${AISA_API_KEY}",
        "api": "openai-completions",
        "models": [

External Transmission

Medium
Category
Data Exfiltration
Confidence
50% confidence
Finding

Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Content

Scanner excerpt · AIsa Chinese Models v3/SKILL.md (reported line 355)May include surrounding context.

md
"models": {
    "providers": {
      "aisa": {
        "baseUrl": "https://api.aisa.one/v1",
        "apiKey": "${AISA_API_KEY}",
        "api": "openai-completions",
        "models": [

External Transmission

Medium
Category
Data Exfiltration
Confidence
50% confidence
Finding

Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Content

Scanner excerpt · AIsa Chinese Models v3/guide-zh-CN.md (reported line 88)May include surrounding context.

md
"models": {
    "providers": {
      "aisa": {
        "baseUrl": "https://api.aisa.one/v1",
        "apiKey": "${AISA_API_KEY}",
        "api": "openai-completions",
        "models": [

External Transmission

Medium
Category
Data Exfiltration
Confidence
50% confidence
Finding

Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Content

Scanner excerpt · SKILL.md (reported line 51)May include surrounding context.

md
"models": {
    "providers": {
      "aisa": {
        "baseUrl": "https://api.aisa.one/v1",
        "apiKey": "${AISA_API_KEY}",
        "api": "openai-completions",
        "models": [

External Transmission

Medium
Category
Data Exfiltration
Confidence
50% confidence
Finding

Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Content

Scanner excerpt · SKILL.md (reported line 219)May include surrounding context.

md
"models": {
    "providers": {
      "aisa": {
        "baseUrl": "https://api.aisa.one/v1",
        "apiKey": "${AISA_API_KEY}",
        "api": "openai-completions",
        "models": [

External Transmission

Medium
Category
Data Exfiltration
Confidence
50% confidence
Finding

Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Content

Scanner excerpt · SKILL.md (reported line 238)May include surrounding context.

md
"models": {
    "providers": {
      "aisa": {
        "baseUrl": "https://api.aisa.one/v1",
        "apiKey": "${AISA_API_KEY}",
        "api": "openai-completions",
        "models": [

External Transmission

Medium
Category
Data Exfiltration
Confidence
50% confidence
Finding

Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Content

Scanner excerpt · SKILL.md (reported line 355)May include surrounding context.

md
"models": {
    "providers": {
      "aisa": {
        "baseUrl": "https://api.aisa.one/v1",
        "apiKey": "${AISA_API_KEY}",
        "api": "openai-completions",
        "models": [

External Transmission

Medium
Category
Data Exfiltration
Confidence
50% confidence
Finding

Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Content

Scanner excerpt · references/config-examples.md (reported line 20)May include surrounding context.

md
"models": {
    "providers": {
      "aisa": {
        "baseUrl": "https://api.aisa.one/v1",
        "apiKey": "${AISA_API_KEY}",
        "api": "openai-completions",
        "models": [

External Transmission

Medium
Category
Data Exfiltration
Confidence
50% confidence
Finding

Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Content

Scanner excerpt · references/config-examples.md (reported line 134)May include surrounding context.

md
"models": {
    "providers": {
      "aisa": {
        "baseUrl": "https://api.aisa.one/v1",
        "apiKey": "${AISA_API_KEY}",
        "api": "openai-completions",
        "models": [

External Transmission

Medium
Category
Data Exfiltration
Confidence
50% confidence
Finding

Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Content

Scanner excerpt · references/guide-zh-CN.md (reported line 88)May include surrounding context.

md
"models": {
    "providers": {
      "aisa": {
        "baseUrl": "https://api.aisa.one/v1",
        "apiKey": "${AISA_API_KEY}",
        "api": "openai-completions",
        "models": [

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
97% confidence
Finding

Passing an API key directly on the command line can expose the secret through shell history, process listings, terminal logging, CI logs, or support screenshots. In a provider-setup skill context, users are likely to copy-paste commands verbatim, which increases the chance of credential leakage and subsequent unauthorized API use.

Content

No source excerpt is available for this finding.

Intent-Code Divergence

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The guide makes absolute privacy assurances such as users 'completely need not worry' and that prompts/replies are 'completely private' based on a zero-data-retention arrangement. Zero retention does not guarantee end-to-end privacy, absence of logging everywhere, lawful access limitations, transport/security correctness, or misuse by intermediaries, so these claims can mislead users into sending sensitive data under a false sense of safety.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
77% confidence
Finding

文件标题表明这是“中文使用指南”，但文档中未说明这是可选的本地化版本，也未提供语言选择或适用范围说明。根据规则，若技能内容强制特定语言而没有用户选择或明确的地区性正当理由，可视为语言/地区策略问题。

Content

No source excerpt is available for this finding.

Missing User Warnings

Low
Category
Not specified by scanner
Confidence
85% confidence
Finding

The curl example sends a bearer token to a third-party endpoint but does not warn that the API key and associated request metadata are being transmitted outside the local environment. While this is expected for an API call, omission of that warning can still mislead users about data handling and increase accidental disclosure risk in copied commands or monitored environments.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.