Model Context Protocol (MCP) server for the manifoldbt backtesting engine
Wellknown found it in public sources; nobody has proven control of it yet. Claiming takes one click if the repository is under your GitHub account, or a small file on your domain otherwise. Verified owners get the badge, 15-minute checks, status alerts, edits that outrank crawled data, and a ranking boost.
Agents can do it too: POST https://wellknown.network/api/v1/claims with {"agent":"manifoldbt-mcp","method":"well_known_file"} — machine-readable steps at claim.json, guide at /docs/claim.
Everything here was measured by our prober or read from a registry. Nothing is self-reported.
Attributed to the source that supplied each field. Treated as claims, not facts.
# manifoldbt-mcp [](https://github.com/ClementG91/manifoldbt-mcp/actions/workflows/ci.yml) [](https://pypi.org/project/manifoldbt-mcp/) [](https://pypi.org/project/manifoldbt-mcp/) [](./LICENSE) > Releasing to PyPI is fully automated via > [GitHub Actions + PyPI Trusted Publishing](./docs/RELEASING.md) — just tag > `vX.Y.Z` and push. > **Model Context Protocol (MCP) server for > [manifoldbt](https://github.com/Jimmy7892/manifoldbt)** — the Rust-powered > backtesting engine for quantitative research. Expose the full manifoldbt backtesting engine to any MCP-compatible client (Claude Desktop, Claude Code, Cursor, Devin, VS Code, Windsurf, …). Let an LLM author strategies, run backtests, sweep parameters, and generate tearsheets — without the user writing a line of Python. This package is **not affiliated with the manifoldbt authors**; it is a thin adapter that imports the published `manifoldbt` wheel and wraps its Python API behind an MCP server. All heavy lifting is done by the underlying Rust engine. --- ## Install ```bash pip install manifoldbt-mcp ``` This pulls in `manifoldbt` and the reference `mcp` Python SDK as dependencies. Python 3.10+ is required. Optional extras: ```bash pip install "manifoldbt-mcp[plot]" # plotly stack, for plot_tearsheet pip install "manifoldbt-mcp[dev]" # pytest, ruff, plotly stack ``` ## Run ```bash manifoldbt-mcp # stdio transport (default) manifoldbt-mcp --http --port 8765 # streamable-HTTP transport manifoldbt-mcp --http --host 0.0.0.0 # remote deployment ``` --- ## Client configuration ### Claude Desktop `~/Library/Application Support/Claude/claude_desktop_config.json` (or `%…
Mapped onto the structured taxonomy from declared text and observed tool names. Confidence shown for derived entries.
Every source is kept verbatim. Field changes are logged as events.