What runtime does VI+ MCP Server require?
Python must be installed and accessible from the terminal. A `.env` file with a `RISK_TOKEN` environment variable is required in the project root.
@jgamblin
基本情報
An MCP Server for Cisco Vulnerability Management VI+ API
以下の設定を使って、このサーバーを MCP 対応クライアントに追加してください。
{
"mcpServers": {
"VI-MCP": {
"command": "docker",
"args": [
"build",
"-t",
"vi_mcp_server",
"."
]
}
}
}ツール一覧はまだありません
この掲載には表示できるツール一覧がまだありません。サーバーにツールがないという意味ではありません。対応機能はプロジェクトのドキュメントをご確認ください。
ツールは README から自動的に抽出されます。メンテナーは ## Tools という見出しの下に記載することで、このタブに反映できます。
The VI+ MCP Server is a Python-based application that fetches and formats details about Common Vulnerabilities and Exposures (CVEs) using the Cisco CVM API, providing structured summaries including risk scores, exploitability, and fixes.
Configure it as an MCP server in VS Code by adding a JSON entry under the "mcp" key in settings.json, specifying the command python and the path to vi-mcp.py. Alternatively, serve it via Open‑WebUI by cloning the repository, building the Docker image, and running the MCP‑to‑OpenAPI proxy (mcpo) with a command that includes the environment file.
Python must be installed and accessible from the terminal. A `.env` file with a `RISK_TOKEN` environment variable is required in the project root.
Use the MCP‑to‑OpenAPI proxy (`mcpo`) with a Docker container: run `uvx mcpo --port 8000 -- docker run -i --rm --env-file .env vi_mcp_server`, then access the auto‑generated OpenAPI documentation at `http://localhost:8000/docs`.
It communicates via stdio when configured as an MCP server in VS Code, and can also be run inside a Docker container for HTTP‑based serving via mcpo.
The README suggests the Gemma 3:12B model (128K context window, multimodal) for optimal performance with MCP in Open‑WebUI.
Run the provided `vi-mcp-test.py` script after configuring the `.env` file. It fetches and displays details for a hardcoded CVE ID (`CVE-2023-35078`).


One case file per task — shared by every session, agent and person that touches it. Self-hosted MCP server and live board. MIT.

Local-first credential broker for MCP. AI coding agents (Claude Code, Cursor, Codex…) call APIs, SSH and databases through scoped presets; secrets stay in your OS keychain and never enter the agent's context.

Verified distributed compute for humans, applications, and agents. Submit Blender, Octane, and video workloads, route them across qualified GPUs, track asynchronous execution, download finished artifacts, and verify FARP

The 9apes MCP server lets an AI assistant read the GitHub Actions data 9apes already collects for your organization: runs, jobs, steps, and their durations. Ask it why a build got slower, which jobs take the longest, or
コメント