Back to skill

Security audit

Mx Financial Assistant

Security checks for vulnerabilities and agentic risk

Overview

This is a coherent financial Q&A skill that calls Eastmoney's remote API, with some privacy and scoping caveats users should understand.

Install only if you are comfortable providing an Eastmoney API key and sending financial questions, which may include investment interests or holdings you type, to Eastmoney's remote service. Avoid including personal identifiers or confidential portfolio details unless that data sharing is acceptable to you.

Vulnerability Patterns
  • Insecure DependenciesIntroduces malicious components through unsafe dependency sources
  • 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)

T08 · Insecure Dependencies

Note
Location
SKILL.md:20
Finding

Unpinned Third-Party Dependency Allows Unreviewed Package Changes

Content
View full analysis

Vulnerability Details

File Location: SKILL.md, lines 20–27
Vulnerability Type: Unpinned third-party dependency
Risk Level: Low

Vulnerable Code

json
{
  "id": "pip-deps",
  "kind": "python",
  "package": "httpx",
  "label": "Install Python dependencies"
}

Technical Analysis

The installation configuration specifies httpx without an exact version constraint or an integrity hash. Dependency resolution can therefore select a different package release each time the Skill is installed.

This does not establish that the current httpx package is malicious. However, it creates a supply-chain exposure because future releases are incorporated without review. A compromised package publisher, package registry, dependency-resolution source, or newly released vulnerable version could cause unsafe code to be installed.

Python package installation may execute package build hooks and other installation-time code. Consequently, a malicious release selected by the unpinned dependency could execute code in the environment performing the installation.

Attack Path

  1. An attacker compromises the relevant package publication channel, registry resolution path, or a future package release.
  2. The attacker publishes or causes resolution to select a malicious or compromised version of httpx.
  3. A user installs the Skill after that version becomes available.
  4. The installer resolves "package": "httpx" to the affected release because no exact version or hash is enforced.
  5. Malicious installation hooks or imported package code execute with the privileges of the installation or runtime process.

Impact Assessment

Successful exploitation could execute arbitrary code with the privileges of the account or service installing or running the Skill. Depending on those privileges, the compromised dependency could access environment variables such as EM_API_KEY, read or modify files available to ...[truncated 309 chars]

Remediation
View remediation

Remediation Suggestions

  1. Pin httpx to an exact, reviewed version rather than allowing unrestricted resolution.
  2. Maintain dependencies in a lock file generated from a trusted environment.
  3. Require cryptographic hashes for downloaded distributions, such as through a hash-locked requirements file.
  4. Install packages only from an explicitly configured, trusted package index.
  5. Regularly scan the pinned version and its transitive dependencies for disclosed vulnerabilities.
  6. Test and review dependency upgrades before changing the pinned version.
  7. Perform installation in a least-privileged, isolated environment without unnecessary credentials or host filesystem access.
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (5)

Vague Triggers

High
Category
Not specified by scanner
Confidence
96% confidence
Finding

The trigger guidance includes very broad everyday phrases such as '请问', '分析一下', and '总结一下', which are not specific to financial tasks. This can cause the skill to be invoked in unrelated contexts, leading to unintended transmission of arbitrary user prompts to an external financial API and incorrect routing of conversations.

Content

No source excerpt is available for this finding.

Undeclared Tool Scope

Medium
Category
MCP Least Privilege
Confidence
87% confidence
Finding

The skill requires an environment secret (EM_API_KEY) and performs networked API access, but it does not declare an explicit tool scope such as permissions or allowed-tools. This weakens least-privilege controls and can allow the runtime to grant broader capabilities than reviewers or orchestrators can easily reason about, increasing the blast radius if the skill is misused or later extended.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
98% confidence
Finding

“始终使用中文进行交互和输出。如用户使用其他语言,仍以中文生成回答内容”构成了明确的语言强制要求。该要求未向用户提供语言选择或 opt-in,也未说明这是仅面向中文地区或合规要求限定的技能,因此符合语言/locale policy violation。

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
90% confidence
Finding

All user-facing natural-language strings, including the tool name, descriptions, prompts, and errors, are written only in Chinese, with no indication that users may choose another language. Under the stated policy, forcing a specific language without user opt-in is a natural-language policy violation unless the locale constraint is clearly documented and justified.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The script sends the user's raw financial question to a third-party remote API, and the code provides no user-facing notice, consent flow, or minimization before transmission. Because financial questions can contain sensitive investment intent, holdings, or personal financial context, this creates a real privacy/security weakness even though HTTPS is used.

Content

No source excerpt is available for this finding.