Vertex AISAFE
MCP server for Vertex AI and Gemini tools, including grounded answers, documentation research, and filesystem workflows.
Overview
From the repository's own README, as read at the audited commit. Badges and raw HTML are left out.
[](https://mseep.ai/app/shariqriazz-vertex-ai-mcp-server)
[](https://smithery.ai/server/@shariqriazz/vertex-ai-mcp-server)
This project implements a Model Context Protocol (MCP) server that provides a comprehensive suite of tools for interacting with Google Cloud's Vertex AI Gemini models, focusing on coding assistance and general query answering.
Security boundary
This server includes filesystem write and terminal-execution tools in addition to model and research tools. Run it with a dedicated low-privilege identity, limit its filesystem and network access, and keep consequential tools behind explicit approval. Do not treat an MCP client confirmation prompt as server-side authorization.
For the implementation tradeoffs this project exposed alongside six other public MCP codebases, see What building seven MCP servers taught me about production MCP. The article uses this repository as public engineering evidence; it does not claim that every production control discussed there is already implemented here.
Features
- Provides access to Vertex AI Gemini models via numerous MCP tools.
- Supports web search grounding (
answer_query_websearch) and direct knowledge answering (answer_query_direct). - Configurable model ID, temperature, streaming behavior, max output tokens, and retry settings via environment variables.
- Uses streaming API by default for potentially better responsiveness.
- Includes basic retry logic for transient API errors.
- Minimal safety
2e8a4b29f159OBSERVED · 2026-10-07Connect
Built from this server's own package name, version and transport as found in its source — not copied from anyone's documentation, so it cannot drift against a page we do not control. Replace the environment placeholders with a token scoped to the least it needs.
claude mcp add vertex-ai-mcp-server --env AI_MAX_OUTPUT_TOKENS=${AI_MAX_OUTPUT_TOKENS} --env GEMINI_API_KEY=${GEMINI_API_KEY} -- npx -y [email protected]{
"mcpServers": {
"vertex-ai-mcp-server": {
"command": "npx",
"args": [
"-y",
"[email protected]"
],
"env": {
"AI_MAX_OUTPUT_TOKENS": "${AI_MAX_OUTPUT_TOKENS}",
"GEMINI_API_KEY": "${GEMINI_API_KEY}"
}
}
}
}Exposed tools (16)
11 read · 5 write · 0 destructive.
| Tool | Risk | Description |
|---|---|---|
answer_query_direct | read | Answers a natural language query using only the internal knowledge of the configured Vertex AI model (${modelIdPlaceholder}). Does not use web search. Requires a |
answer_query_websearch | read | Answers a natural language query using the configured Vertex AI model (${modelIdPlaceholder}) enhanced with Google Search results for up-to-date information. Requires a |
code_analysis_with_docs | read | Analyzes code snippets by comparing them with best practices from official documentation found via web search. Identifies potential bugs, performance issues, and security vulnerabilities. Uses the configured Vertex AI model (${modelIdPlaceholder}) with Google Search. Requires |
database_schema_analyzer | read | Reviews database schemas for normalization, indexing, and performance issues. Suggests improvements based on database-specific best practices. Provides migration strategies for implementing suggested changes. Uses the configured Vertex AI model (${modelIdPlaceholder}) with Google Search. Requires |
dependency_vulnerability_scan | read | Analyzes project dependencies for known security vulnerabilities. Provides detailed information about each vulnerability with severity ratings. Suggests mitigation strategies and secure alternatives. Uses the configured Vertex AI model (${modelIdPlaceholder}) with Google Search. Requires |
documentation_generator | read | Creates comprehensive documentation for code, APIs, or systems. Follows industry best practices for technical documentation. Includes examples, diagrams, and user guides. Uses the configured Vertex AI model (${modelIdPlaceholder}) with Google Search. Requires |
get_doc_snippets | read | Provides precise, authoritative code snippets or concise answers for technical queries by searching official documentation. Focuses on delivering exact solutions without unnecessary explanation. Uses the configured Vertex AI model (${modelIdPlaceholder}) with Google Search. Requires |
microservice_design_assistant | read | Helps design microservice architectures for specific domains. Provides service boundary recommendations and communication patterns. Includes deployment and orchestration considerations. Uses the configured Vertex AI model (${modelIdPlaceholder}) with Google Search. Requires |
regulatory_compliance_advisor | read | Provides guidance on regulatory requirements for specific industries (GDPR, HIPAA, etc.). Suggests implementation approaches for compliance. Includes checklists and verification strategies. Uses the configured Vertex AI model (${modelIdPlaceholder}) with Google Search. Requires |
save_answer_query_direct | write | Answers a natural language query using only the internal knowledge of the configured Vertex AI model (${modelIdPlaceholder}), does not use web search, and saves the answer to a file. Requires |
save_answer_query_websearch | write | Answers a natural language query using Google Search results and saves the answer to a file. Uses the configured Vertex AI model (${modelIdPlaceholder}). Requires |
save_doc_snippet | write | Provides precise code snippets or concise answers for technical queries by searching official documentation and saves the result to a file. Uses the configured Vertex AI model (${modelIdPlaceholder}) with Google Search. Requires |
save_generate_project_guidelines | write | Generates comprehensive project guidelines based on a tech stack using web search and saves the result to a specified file path. Uses the configured Vertex AI model (${modelIdPlaceholder}). Requires |
save_topic_explanation | write | Provides a detailed explanation for a query about a specific software topic using official documentation found via web search and saves the result to a file. Uses the configured Vertex AI model (${modelIdPlaceholder}). Requires |
security_best_practices_advisor | read | Provides security recommendations for specific technologies or scenarios. Includes code examples for implementing secure practices. References industry standards and security guidelines. Uses the configured Vertex AI model (${modelIdPlaceholder}) with Google Search. Requires |
testing_strategy_generator | read | Creates comprehensive testing strategies for applications or features. Suggests appropriate testing types (unit, integration, e2e) with coverage goals. Provides example test cases and testing frameworks. Uses the configured Vertex AI model (${modelIdPlaceholder}) with Google Search. Requires |
Trust audit
SAFEgrade B · trust 89/100 Nothing in the source contradicts what it says it does. Grade A is reserved for packages that have also passed the behavioural sandbox.
| Layer | What it checks | Result |
|---|---|---|
| L0 | Provenance & inventory | PASS |
| L1 | Static analysis of the code | PASS |
| L2 | Instruction surface (what it tells the agent) | PASS |
| L3 | Class-specific surface | PASS |
| L4 | Behavioural (sandbox) | SKIPPED |
What the source does
- Filesystem
- none-observed
- Network
- none-observed
- Shell
- none-observed
- Dependencies
- not all pinned
- Secrets in source
- none-found
Findings (2)
@google/genai, diff, dotenv, minimatch, zod, zod-to-json-schema, @types/diff, @types/minimatch
Minimal safety filters applied (`BLOCK_NONE`) to reduce potential blocking (use with caution).
Gates applied: no_behavioural_pass.
2e8a4b29f159full audit observations/trust-audit/mcp-server/shariqriazz__vertex-ai.json · Report an issue / request a re-scanAudit history
Every audit this server has had. A grade with a past is a grade somebody is still checking.
| Date | Source | Verdict | Grade | Score | Change |
|---|---|---|---|---|---|
| 2026-10-07 | 2e8a4b29f159 | SAFE | B | 89 | first audit |
Questions
What is the Vertex AI MCP server?
MCP server for Vertex AI and Gemini tools, including grounded answers, documentation research, and filesystem workflows.
What tools does Vertex AI expose?
16 in total: 11 read-only, 5 that write, and 0 that can delete or overwrite. Every one is listed on this page with its risk.
Is Vertex AI safe to connect to an agent?
The audit found nothing in the source that contradicts what it says it does, and graded it B (89/100). Grade A is held back for packages that have also passed a sandboxed behavioural run, which is why a clean server reads B.
What credentials does Vertex AI need?
It reads AI_MAX_OUTPUT_TOKENS and GEMINI_API_KEY from the environment. Give it a token scoped to the least it needs — an agent that can be talked into calling a tool can be talked into calling it with your credentials.
How does Vertex AI run?
It speaks stdio, so it runs as a local process your client starts. It is published on npm as vertex-ai-mcp-server at 0.4.0.
How current is this page?
The grade is for one exact copy of the source (2e8a4b29f159), read on 2026-10-07. The repository is watched and re-audited when it changes.