Security audit

Alibaba Cloud AI Search DashVector

Security checks for vulnerabilities and agentic risk

Overview

This skill is a straightforward DashVector integration, but users should treat its API key and endpoint configuration carefully.

Install the SDK in an isolated virtual environment, consider pinning the `dashvector` package version, and only run the quickstart with a least-privileged API key and a verified DashVector cluster endpoint because it can create a collection and upsert sample documents.

A.I.G

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 (2)

T08 · Insecure Dependencies

Warning

Location: SKILL.md:18
Finding: Unpinned DashVector Dependency Creates Supply-Chain Exposure
Content: View full analysis
Remediation: View remediation
``` 2. Store dependencies in a lock file or hashed requirements file. 3. Require cryptographic hashes during installation: ```bash python -m pip install --require-hashes -r requirements.txt ``` 4. Generate and review hashes from a trusted package source before publishing the skill. 5. Document the expected official package index and avoid untrusted extra indexes. 6. Use automated dependency scanning and controlled update procedures so version changes receive review before distribution. 7. Install and execute the package in an isolated, least-privileged virtual environment. ]]>

T09 · Insecure Skill Coding Practices

Warning

Location: scripts/quickstart.py:27
Finding: Environment-Controlled Endpoint May Receive the DashVector API Key
Content: View full analysis
Remediation: View remediation

SkillSpector

Vulnerability Patterns

MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code

Findings (1)

Undeclared Tool Scope

Medium

Category: MCP Least Privilege
Confidence: 86% confidence
Finding: The skill references sensitive environment variables (DASHVECTOR_API_KEY, DASHVECTOR_ENDPOINT) and provides executable commands, but it does not declare any explicit tool scope such as permissions or allowed-tools. In an agent environment, this can lead to overbroad access to environment secrets or unintended execution pathways because the runtime is not constrained to only the minimum capabilities needed.
Content: No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.