Back to skill

Security audit

Maxhub Lemon8

Security checks for vulnerabilities and agentic risk

Overview

ClawScan marks this skill suspicious because its Lemon8 workflow includes instructions to route some requests to Douyin APIs; the scan is partial, so the evidence is incomplete.

Consider the suspicious verdict before installing. The scan recommends not installing and is incomplete: six of seven components were fully inspected, while one was partially inspected. Review the Douyin fallback and the external data flow, and note that the install command does not pin the package version.

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
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • Output HandlingUnvalidated Output Injection, Cross-Context Output, Unbounded Output
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
Findings (8)

Incomplete referenced artifact analysis

High
Category
analysis-evasion
Confidence
100% confidence
Finding

Referenced artifact was not completely inspected

Content

Scanner excerpt · SKILL.md (reported line 132)May include surrounding context.

md
, comment, follower, following, topic, information, list, profile, ids, item | `references/api-post-user.md` | fetch_discover_tab_information_tabs, fetch_post_d

MCP server referenced without pinned version: 'npx clawhub'.

Medium
Category
MCP Rug Pull
Confidence
91% confidence
Finding

The installation command uses npx to resolve and execute an unpinned clawhub package, exposing installation to unexpected or compromised future releases. Although this is a CLI installer rather than an MCP server, a malicious resolved version could execute with the installing user's permissions.

Content

No source excerpt is available for this finding.

Example invocation prompts use generic words without requiring Lemon8-specific context.

Medium
Category
Trigger Abuse
Confidence
95% confidence
Finding

The example prompts include 搜索 (search), 用户 (user), 分享 (share), and 链接 (link), which commonly occur in requests unrelated to Lemon8. Although the README identifies the skill as a Lemon8 assistant, these invocation examples do not require that context or provide exclusions, creating ambiguity about when the skill should activate.

Content

No source excerpt is available for this finding.

MCP server referenced without pinned version: 'npx clawhub'.

Medium
Category
MCP Rug Pull
Confidence
91% confidence
Finding

The documented npx command executes an unpinned clawhub package, so installation behavior can change when the registry release changes. This is a CLI supply-chain risk rather than an MCP configuration issue, and a compromised resolved release could execute arbitrary code with the user's permissions.

Content

No source excerpt is available for this finding.

Example invocation commands are broad everyday terms without Lemon8-specific scope.

Medium
Category
Trigger Abuse
Confidence
95% confidence
Finding

The example commands include generic terms such as 搜索 (search), 用户 (user), 分享 (share), and 链接 (link). Although the README identifies the skill as a Lemon8 assistant, these invocation examples lack platform qualifiers or exclusion conditions, making them ambiguous for unrelated search, user, or link requests.

Content

No source excerpt is available for this finding.

External Transmission

Medium
Category
Data Exfiltration
Confidence
60% 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 53)May include surrounding context.

md
maxhub_auth_header="Authorization: Bearer ${MAXHUB_API_KEY}"

# GET example
curl -s "https://www.aconfig.cn/api/v1/lemon8/{endpoint}?%7Bparams=}" \
  -H "$maxhub_auth_header"

# POST example

The autonomous query workflow lacks a privacy warning about sending user-supplied search terms and target identifiers to MaxHub.

Medium
Category
Output Handling
Confidence
86% confidence
Finding

The workflow instructs the agent to search for users and retrieve their information, executing planned queries autonomously through a third-party API. Although the document identifies MaxHub and includes a data-accuracy disclaimer, it does not warn that search terms and target identifiers supplied by the user will be transmitted to that service.

Content

No source excerpt is available for this finding.

The fallback routing table directs requests to Douyin APIs despite the skill's Lemon8-only purpose.

Medium
Category
Excessive Agency
Confidence
94% confidence
Finding

The fallback table authorizes Douyin requests inside a Lemon8-only workflow, creating a scope-control defect that could cause unintended cross-platform queries, charges, or misleading results if followed. Its source endpoints are also Douyin-specific, so normal Lemon8 calls do not clearly reach these mappings; this limits demonstrated exploitability but does not make the erroneous operational instructions safe.

Content

No source excerpt is available for this finding.