FlowProof
FlowProof is an MCP server that enables AI assistants to reproducibly execute bioinformatics pipelines with verifiable provenance. You can:
List pipelines: Retrieve registered pipelines with IDs, descriptions, and read types (e.g., short-read, ONT long-read).
Describe a pipeline: Get detailed inputs, parameters, and outputs for a specific pipeline.
Run a pipeline: Launch a pipeline with inputs and parameters, receiving a run ID for tracking.
Check run status: Query the current state (running, completed, failed) of a run.
Retrieve results: Obtain a manifest of output files and their SHA-256 checksums.
Get provenance: Download a Workflow Run RO-Crate record capturing workflow version, container digests, tool versions, exact parameters, and all input/output checksums for independent verification.
Enables execution of bioinformatics pipelines using Nextflow as the workflow backend, providing reproducible run management and verifiable provenance.
Click on "Deploy 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., "@FlowProof run the ONT assembly on sample.fastq"
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.
FlowProof
Reproducible bioinformatics pipeline execution over the Model Context Protocol (MCP), with verifiable provenance.
FlowProof lets an AI assistant run a bioinformatics pipeline and hand back results whose provenance can be independently verified: pipeline version, container digests, tool versions, parameters, and SHA-256 checksums of every input and output, emitted as a Workflow Run RO-Crate.
It fills the layer the MCP-for-bioinformatics ecosystem is missing. Data access already has MCP servers (BioMCP); analysis planning already has AI agents (Biomni, AutoBA). The gap is reliable, trustworthy execution, which needs orchestration and provenance discipline. That is what FlowProof provides.
Why it exists
An AI that "runs an analysis for you" is only useful if you can trust the result. FlowProof makes every run reproducible and independently checkable, so an AI-driven result is not a black box: it ships with the exact recipe and checksums to reproduce it byte-for-byte.
Related MCP server: SciAgentKit
Status
FlowProof is working and installable today (pip install flowproof-mcp). The pipeline registry, execution backends, and RO-Crate provenance are complete and covered by tests, and the MCP server exposes six tools (list, describe, run, status, results, provenance). It ships with two execution backends: a zero-dependency backend for development and CI, and a Nextflow backend that runs real pipelines.
Install
uv syncUsing FlowProof
FlowProof works two ways. Both let an AI assistant run pipelines for you; you never touch a terminal after setup.
Local (recommended for real data)
Runs on your own machine, so your data never leaves it and your compute runs the pipelines. No token needed.
Add this to your MCP client config (Claude Desktop: claude_desktop_config.json; Cursor: MCP settings):
{
"mcpServers": {
"flowproof": {
"command": "uvx",
"args": ["flowproof-mcp"]
}
}
}Then just ask your assistant: "list the FlowProof pipelines" or "run the ONT assembly on this file". Runs are written under ~/.flowproof/runs (override with FLOWPROOF_RUNS_DIR).
Cloud (instant, no install)
Connect your client to the hosted server. Nothing to install; the cloud runs it. Uses a bearer token today (per-user keys and OAuth are on the roadmap):
{
"mcpServers": {
"flowproof": {
"url": "https://flowproof.specvista.com/mcp/",
"headers": { "Authorization": "Bearer YOUR_TOKEN" }
}
}
}Run the server directly
flowproof # stdio (local, default)
FLOWPROOF_TRANSPORT=http flowproof # HTTP server on :8000Tools
Tool | Purpose |
| Registered pipelines with id, description, read type |
| Inputs, parameters, outputs for a pipeline |
| Execute a pipeline reproducibly, returns a run id |
| Status of a run |
| Output file manifest with checksums |
| The verifiable Workflow Run RO-Crate record |
Pipelines (seed)
id | Read type | Analysis |
| long | Oxford Nanopore read statistics (QC); optional Flye assembly. Runs on a built-in sample if you provide no input. |
| short | Short-read QC to germline variant calling (nf-core/sarek) |
| short | RNA-seq quantification: QC, trimming, alignment, gene counts (nf-core/rnaseq) |
| short | Fetch raw reads and metadata from public archives, SRA/ENA/GEO (nf-core/fetchngs) |
| short | Viral genome reconstruction and variant calling from amplicon data (nf-core/viralrecon) |
New pipelines register by manifest; the server does not change.
Try it instantly: ask your assistant to "run ont-read-stats" with no input, and FlowProof runs the bundled Oxford Nanopore sample end to end and returns a verifiable provenance crate, no data or reference genome required.
Backends
MockBackend: deterministic, dependency-free. Used for development and tests.NextflowBackend: runsnextflow runwith a container profile. Requires Nextflow and Docker.
Provenance
Every run emits ro-crate-metadata.json following the Workflow Run RO-Crate profile, capturing the workflow and version, container images, resolved tool versions, exact parameters, and SHA-256 checksums of all inputs and outputs.
Develop
PYTHONPATH=src uv run --with pytest --no-project python -m pytest tests/ -qArchitecture
See DESIGN.md.
Cite
If you use FlowProof in your research, please cite the archived software release:
Ajibade, H. A. (2026). FlowProof: reproducible bioinformatics pipeline execution over the Model Context Protocol with verifiable provenance (v0.1.2). Zenodo. https://doi.org/10.5281/zenodo.21932977
A machine-readable CITATION.cff is included, so GitHub shows a "Cite this repository" button with BibTeX and APA formats.
@software{flowproof,
author = {Ajibade, Hammed Adedapo},
title = {FlowProof: reproducible bioinformatics pipeline execution over the Model Context Protocol with verifiable provenance},
year = {2026},
version = {0.1.2},
publisher = {Zenodo},
doi = {10.5281/zenodo.21932977},
url = {https://github.com/ajibadedapo/flowproof-mcp}
}Available Tools
6 toolsdescribe_pipelineA
Return the full manifest for one pipeline: its required inputs, tunable parameters, container image, and output file patterns. Call this before run_pipeline to learn exactly which inputs and params it expects.
Inputs: pipeline_id (required).
get_provenanceA
Get the Workflow Run RO-Crate provenance for a run: workflow version, container digests, tool versions, parameters, and input/output SHA-256 checksums, the full record needed to reproduce the run byte-for-byte.
Inputs: run_id (required).
get_resultsA
Get the outputs of a run: each output file with its SHA-256 checksum and size in bytes, plus the run status. Use the checksums to verify results independently.
Inputs: run_id (required).
get_run_statusA
Get the current status of a run (queued, running, succeeded, or failed) by its run_id, as returned by run_pipeline.
Inputs: run_id (required).
list_pipelinesA
List the available bioinformatics pipelines. Returns each pipeline's id, a short description, and its read_type (short-read or long-read). Call this first to discover what can be run, then describe_pipeline for details.
Inputs: no named parameters. Returns: result
run_pipelineA
Start a pipeline run and return a run_id to track it. Provide inputs (a map of input name to file path or value) and params (a map of parameter name to value), as described by describe_pipeline. The run executes asynchronously; poll get_run_status, then fetch get_results and get_provenance when it completes.
Inputs: pipeline_id (required), inputs, params.
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
TDQS
Scored across 6 tools
Each tool has a distinct purpose: listing, describing, running, status checking, result retrieval, and provenance. No two tools overlap in function, so an agent can easily select the correct one.
All tool names follow a consistent verb_noun pattern (list_pipelines, describe_pipeline, run_pipeline, get_run_status, get_results, get_provenance). The style is uniform and predictable.
Six tools is well-scoped for a pipeline management server, covering discovery, execution, monitoring, results, and provenance without excess or redundancy.
The workflow from listing pipelines to retrieving results and provenance is fully covered. A cancel/abort run operation is missing, but the core lifecycle is otherwise complete.
Maintenance
Related MCP Connectors
MCP Hub: AI service discovery, per-user OAuth, and multi-service workflow orchestration
Create, browse, remix, collaborate on, and run durable AI workflow nodes from MCP hosts.
AI-powered bioprotocol optimization — generate, search, and manage lab protocols via MCP
Build, test, publish, and run Kovca workflows from AI agents over remote MCP.
Related MCP Servers
- AlicenseNot gradedqualityCmaintenanceExposes pipen bioinformatics pipelines as MCP tools, allowing AI assistants to discover and run complex workflows through a progressive disclosure interface.1MIT
- AlicenseNot gradedqualityCmaintenanceMCP-native scientific skills for reproducible computational biology and AI-driven drug-discovery workflows. It combines deterministic scientific tools with an MCP server to give AI agents real computational capabilities.Apache 2.0
- AlicenseNot gradedqualityCmaintenanceEnables AI-assisted scientific research workflow management through MCP, including project creation, ideation, experiment execution, and artifact handling, with integration for ChatGPT, Codex, and Claude Code.Apache 2.0
- AlicenseBqualityCmaintenanceEnables biomedical research workflows via MCP, including experiment and document search, PDF text extraction, evidence retrieval, and validation with provenance tracking.10MIT