This server exposes Bazel build system capabilities to AI assistants via the Model Context Protocol (MCP), allowing you to query, build, test, and analyze Bazel workspaces.
bazel_query: Run arbitrary Bazel query expressions with configurable output formats (label, build, xml, package, location, graph, etc.)
list_targets: List all targets in a package, subtree, or entire workspace using patterns like //... or //pkg/...
get_deps: Retrieve direct or transitive dependencies of a target up to a configurable depth
get_rdeps: Find all targets that depend on a given target within a specified scope
show_target_info: Display the BUILD rule definition for a target using bazel query --output=build
find_affected_targets: Identify which targets are affected by changes to specific files
bazel_build: Run bazel build on one or more targets with optional flags, returning stdout, stderr, and exit code with structured failure diagnostics
bazel_test: Execute bazel test and get structured pass/fail results including failure log excerpts and duration per test target
explain_build_file: Read and return the contents of a BUILD file (by path relative to workspace root) for analysis
Provides tools for interacting with the Bazel build system, enabling AI assistants to run queries, list targets, get dependencies, perform builds and tests, and analyze BUILD files.
A Model Context Protocol server that exposes Bazel build system capabilities to AI assistants.
Repo: github.com/saeid-rez/bazel-mcp
Tool | Description |
| Run arbitrary |
| List targets in a package or subtree |
| Direct or transitive dependencies of a target |
| Reverse dependencies within a scope |
| Rule definition via |
| Find targets affected by changed files |
| Run |
| Run |
| Read BUILD file contents for analysis |
Related MCP server: mcp-cli-catalog
Requires uv. Run against a Bazel workspace. Open that repo in Cursor, then add to .cursor/mcp.json.
{
"mcpServers": {
"bazel": {
"command": "uvx",
"args": ["bazel-mcp-server"]
}
}
}{
"mcpServers": {
"bazel": {
"command": "uvx",
"args": ["--from", "git+https://github.com/saeid-rez/bazel-mcp", "bazel-mcp-server"]
}
}
}Bazel commands run in your open workspace by default.
If your MCP client does not start the server from the Bazel workspace, pass the workspace explicitly:
{
"mcpServers": {
"bazel": {
"command": "uvx",
"args": [
"bazel-mcp",
"--workspace-root",
"/path/to/your/bazel/workspace"
]
}
}
}Other optional flags:
bazel-mcp-server --bazel-path bazelisk --timeout 600 --max-output-chars 64000Clone the repo and run tests locally:
git clone https://github.com/saeid-rez/bazel-mcp
cd bazel-mcp
uv sync --group dev
uv run pytestMCP server exposing the Backtest360 engine API as tools for AI agents.
A comprehensive Model Context Protocol (MCP) server that enables AI assistants to interact with yo…
Local-first RAG engine with MCP server for AI agent integration.
We provide all the information about MCP servers via our MCP API.
curl -X GET 'https://glama.ai/api/mcp/v1/servers/saeid-rez/bazel-mcp'
If you have feedback or need assistance with the MCP directory API, please join our Discord server