Click on "Install Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type @ followed by the MCP server name and your instructions, e.g., "@codeql-lsp-mcp Write a CodeQL query to find all eval calls in JavaScript."
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
A Model Context Protocol (MCP) server that wraps the CodeQL Language Server Protocol (LSP) to enable LLM agents to write CodeQL queries with intelligent code completion, hover information, and other language features. This server reuses key concepts from the VS Code CodeQL extension to provide a robust language server interface. This is different from a MCP server that wraps the CodeQL CLI tool. However, you can also add CodeQL CLI commands to the server too. Feel free to contribute.
CodeQL CLI installed and available in PATH or set CODEQL_PATH environment variable
Node.js 18 or later
cd codeql-lsp-mcp
npm install
npm run buildAdd to your MCP client configuration:
{
"mcpServers": {
"codeql": {
"command": "node",
"args": ["/path/to/codeql-lsp-mcp/dist/index.js"]
}
}
}[mcp_servers.codeql]
command = "node"
args = ["/path/to/codeql-lsp-mcp/dist/index.js"]If you would like to enable logging, in the LSP client you can set this.verbose = true;.
Related MCP server: LSP-MCP
codeql_open_fileOpen a CodeQL file in the language server.
{
"tool": "codeql_open_file",
"arguments": {
"file_uri": "file:///workspace/query.ql",
"content": "import javascript\n\nfrom Function f\nwhere f.getName() = \"eval\"\nselect f"
}
}codeql_completeGet code completions at a specific position.
{
"tool": "codeql_complete",
"arguments": {
"file_uri": "file:///workspace/query.ql",
"line": 2,
"character": 5,
"trigger_character": "."
}
}codeql_hoverGet hover information (documentation) at a position.
{
"tool": "codeql_hover",
"arguments": {
"file_uri": "file:///workspace/query.ql",
"line": 1,
"character": 7
}
}codeql_definitionGo to definition for a symbol.
{
"tool": "codeql_definition",
"arguments": {
"file_uri": "file:///workspace/query.ql",
"line": 3,
"character": 10
}
}codeql_diagnosticsGet diagnostics (errors, warnings) for a file.
{
"tool": "codeql_diagnostics",
"arguments": {
"file_uri": "file:///workspace/query.ql"
}
}codeql_formatFormat a CodeQL file or selection.
{
"tool": "codeql_format",
"arguments": {
"file_uri": "file:///workspace/query.ql",
"range": {
"start": { "line": 0, "character": 0 },
"end": { "line": 10, "character": 0 }
}
}
}codeql_referencesFind all references to a symbol at a specific position.
{
"tool": "codeql_references",
"arguments": {
"file_uri": "file:///workspace/query.ql",
"line": 3,
"character": 10
}
}codeql_update_fileUpdate the content of an open file.
{
"tool": "codeql_update_file",
"arguments": {
"file_uri": "file:///workspace/query.ql",
"content": "import javascript\n\nfrom Function f\nwhere f.getName() = \"eval\"\nselect f, \"Dangerous eval usage\""
}
}codeql_set_workspaceSet workspace folders for better CodeQL analysis.
{
"tool": "codeql_set_workspace",
"arguments": {
"folders": ["/path/to/codeql/libraries", "/path/to/project"]
}
}This MCP server was developed as part of the framework for our paper QLCoder. Consider citing our paper.
@misc{wang2025qlcoderquerysynthesizerstatic,
title={QLCoder: A Query Synthesizer For Static Analysis of Security Vulnerabilities},
author={Claire Wang and Ziyang Li and Saikat Dutta and Mayur Naik},
year={2025},
eprint={2511.08462},
archivePrefix={arXiv},
primaryClass={cs.CR},
url={https://arxiv.org/abs/2511.08462},
}This server cannot be installed
Unclaimed servers have limited discoverability.
If you are the server author, to access and configure the admin panel.
Code intelligence for coding agents: semantic, AST, graph, and full-text search. 279+ languages.
The team layer for AI coding agents: shared contracts, collision alerts, E2EE sessions.
Cross-agent artifact workspace with provenance across Claude Code, Codex, Cursor, LangGraph.
We provide all the information about MCP servers via our MCP API.
curl -X GET 'https://glama.ai/api/mcp/v1/servers/neuralprogram/codeql-lsp-mcp'
If you have feedback or need assistance with the MCP directory API, please join our Discord server