Back to skill

Security audit

Maxhub Temp Mail

Security checks for vulnerabilities and agentic risk

Overview

The skill is rated suspicious because scanners report a mismatch with its temporary-email purpose, broad request routing, unrelated fallback routes, and an unpinned installation command; the evidence is incomplete.

Consider holding off on installation until the purpose mismatch and unrelated routes are clarified, and the package version can be pinned to a reviewed release. The scan is partial: it inspected all six listed components but marked analysis incomplete because two references were unresolved, so this evidence cannot establish that unreviewed referenced material is safe.

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
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Output HandlingUnvalidated Output Injection, Cross-Context Output, Unbounded Output
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
Findings (9)

Description-Behavior Mismatch

High
Category
MCP Tool Poisoning
Confidence
94% confidence
Finding

The supplied chunk implements an API-key presence check rather than the declared temporary-email functionality. Such a check could support a larger implementation, but no connection to a temporary-email service or implementation of the advertised operations is shown. It does not reveal the key value or transmit data.

Content

No source excerpt is available for this finding.

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

Medium
Category
MCP Rug Pull
Confidence
93% confidence
Finding

The installation command can download and execute an unpinned version of clawhub, allowing a compromised future release to run with the installing user's privileges. This is an installation supply-chain risk, although the finding incorrectly describes the CLI as an MCP server.

Content

No source excerpt is available for this finding.

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

Medium
Category
MCP Rug Pull
Confidence
93% confidence
Finding

The npx installation example does not pin the clawhub package version, so installation may execute newly published code that was never reviewed with this skill. A compromised release could access local credentials or alter files; this is a CLI supply-chain issue rather than an MCP-server configuration issue.

Content

No source excerpt is available for this finding.

Invocation examples use generic email terms without limiting activation to temporary-email requests.

Medium
Category
Trigger Abuse
Confidence
94% confidence
Finding

The example commands “邮箱,邮件” mean “mailbox, email” and can apply to ordinary email tasks, not just temporary-email services. Although the introduction describes a temporary-email assistant, these invocation examples provide no qualifiers or exclusions to distinguish temporary-mail requests from requests about an existing personal or work mailbox.

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/temp-mail/{endpoint}?%7Bparams=}" \
  -H "$maxhub_auth_header"

# POST example

Routing triggers include everyday words and default ambiguous requests to email operations.

Medium
Category
Trigger Abuse
Confidence
95% confidence
Finding

The routing rules match extremely broad terms such as 'get' and explicitly default uncertain requests to email operations. Combined with autonomous query execution, this can turn ambiguous requests into unintended third-party API calls, potentially disclosing supplied parameters or consuming quota.

Content

No source excerpt is available for this finding.

Registration workflows omit privacy warnings about third-party handling of verification emails.

Medium
Category
Output Handling
Confidence
90% confidence
Finding

The skill promotes temporary mail for privacy-preserving registration and automated verification without explaining that the mailbox provider handles sensitive verification messages. Users may mistake protection from exposing their personal address for confidentiality from the provider, placing verification codes or recovery links in a mailbox with unspecified access and retention guarantees.

Content

No source excerpt is available for this finding.

The skill includes operational fallback routes for Douyin video retrieval, search, and user-profile queries, which are unrelated to temporary email.

Medium
Category
Excessive Agency
Confidence
94% confidence
Finding

The operational fallback table authorizes authenticated requests to unrelated Douyin video, search, and profile endpoints, expanding the skill beyond its temporary-email purpose. If the agent follows these routes, it could make unintended queries, transmit query parameters, or consume API quota, although the supplied text does not prove these conditional routes will be reached.

Content

No source excerpt is available for this finding.

The example activation prompts are generic email terms rather than temporary-email-specific requests.

Low
Category
Trigger Abuse
Confidence
95% confidence
Finding

The prompts “邮箱” (mailbox) and “邮件” (email) overlap with ordinary requests about existing email accounts. They do not distinguish temporary-email operations from general email assistance, and no exclusion conditions clarify that boundary.

Content

No source excerpt is available for this finding.