Back to skill

Security audit

DolphinDB 量化金融技能

Security checks for vulnerabilities and agentic risk

Overview

This looks like a real DolphinDB quant-finance helper, but it requires unverified external shell setup and repeats default admin database credentials, so users should review it carefully before installing.

Install only if you trust the referenced DolphinDB setup skills and can review or pin the shell scripts they source. Replace the sample admin credentials with a least-privilege account and secrets-based configuration before adapting any examples, and avoid exposing DolphinDB port 8848 to untrusted networks.

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 (3)

T08 · Insecure Dependencies

Error
Location
SKILL.md:8
Finding

Mandatory Execution of Unreviewed External Shell Scripts

Content
View full analysis
source ../dolphindb-skills/scripts/dolphin_wrapper.sh # Method 2: Run from any location source ~/.jvs/.openclaw/workspace/skills/dolphindb-skills/scripts/dolphin_global.sh ``` A second external script is sourced later: ```bash # Load the environment detector source ../dolphindb-skills/scripts/load_dolphindb_env.sh ``` ### Technical Analysis The skill declares environment initialization to be mandatory and directs the user or agent to execute several shell scripts through `source`. These scripts are not included in the audited project, so their behavior and integrity cannot be verified from the available artifact. Unlike launching an isolated subprocess, `source` evaluates the target script in the current shell. The external script consequently inherits the invoking process's permissions and can: - Execute arbitrary commands with the invoking user's privileges. - Read or alter environment variables and shell state. - Define or replace functions, aliases, and command wrappers. - Change the working directory or shell options. - Access files and credentials available to the invoking process. - Influence all subsequent commands in the same shell session. The referenced paths depend on mutable sibling installations. No exact dependency version, cryptographic digest, ownership validation, or integrity check is required before execution. ### Attack Path 1. An attacker compromises, replaces, or modifies the installed `dolphindb-skills` dependency or one of its referenced shell scripts. 2. A user or agent loads this skill and follows its mandatory initialization procedure. 3. The current shell executes the attacker-controlled script through `so ...[truncated 685 chars]
Remediation
View remediation

T08 · Insecure Dependencies

Warning
Location
SKILL.md:32
Finding

Unpinned Third-Party Skill Installation

Content
View full analysis
Remediation
View remediation

T09 · Insecure Skill Coding Practices

Error
Location
SKILL.md:64
Finding

Hardcoded Default DolphinDB Administrative Credentials

Content
View full analysis
Remediation
View remediation
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (3)

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
74% confidence
Finding

SQP-3 适用于所有文件类型,且可检查自然语言中的语言/区域策略违规。文档以“强制流程”“无论在何种场景下调用此技能”这类绝对化措辞要求统一流程,并整体仅以中文提供使用说明与触发条件,没有声明可根据用户偏好切换语言,也没有说明必须中文的合规或地域性理由。

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
99% confidence
Finding

The skill repeatedly shows connecting as admin with the default-looking password "123456" and provides no warning to replace it or use secrets management. In practice, users often copy examples verbatim, which can normalize insecure deployments and enable unauthorized database access if the same credentials are active in a reachable environment.

Content

No source excerpt is available for this finding.

Intent-Code Divergence

Low
Category
Not specified by scanner
Confidence
77% confidence
Finding

Lines L057-L083 state that all DolphinDB scripts in Python should be invoked through s.run(), presenting a unified calling approach. However, the same document also requires shell source commands for environment setup and uses Python-side s.connect()/s.close() calls, so the wording overstates what actually happens and can mislead readers about execution boundaries.

Content

No source excerpt is available for this finding.