Runs a SonarQube scan for a project using sonar-scanner-cli and the configured SonarQube server, requiring a sonar-project.properties file in the project.
This server does not currently have a deployable release on Glama. Follow the repository's setup instructions to install and configure it for your MCP client, including any required credentials and local dependencies.
An MCP server that let an LLM run a project's real test suite and SonarQube scan in throwaway containers, without a raw shell.
Archived. Stopped. The README says what it tried and what survived.
Archived 2026-11-01. Nothing called it after the migration check on 2026-10-31, and chargehand now runs the real tests.
Whether an agent could run a repository's own tests and quality scan through four narrow MCP tools
(list_projects, run_tests, sonar_scan, sonar_quality_gate) instead of a shell. Each run happened in a
one-shot container: the repo mounted read-only, a tmpfs for scratch space, and Docker reached only through a
socket proxy limited to containers, POST and images.
Related MCP server: Docker MCP Server
I wrote it on 2026-09-28 and tagged seven releases, 0.1.0 to 0.7.0, by 2026-09-29. The same day I removed it from my
Claude clients because its auth was broken and I barely used it. It kept running behind my MCP gateway for a month.
On 2026-11-01 I read the gateway log for the last 30 days: 6 requests, all initialize or tools/list from a
migration smoke test, and no tools/call. The agent that really needed test runs, chargehand, got its own.
The socket-proxy pattern: docs/guide/security-model.md says why a mounted Docker socket is root on the host and which proxy flags narrow it.
A project allowlist with per-project limits, and a cap on concurrent runs.
The full working README and the user guide: docs/history.md, docs/guide/.
Test runs belong to chargehand. The stack was removed from my server in Egoushka/homelab#178; this repo stays readable under the MIT licence.
Made by Yehor Hrabovskyi.
This server cannot be deployed
Operate Linux, macOS and Windows from your LLM. Every action runs through an auditable allowlist.
Build, test and cache insights for Xcode, Gradle and Bazel projects, for coding agents.
Run AI evaluations, record agent traces and gate releases on evidence, with scoped org access.
Run verified read-only code tools: quant diagnostics + agent-ops preflight, no source exposure.
