More screenshots

    Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIMuseMuseDotsDots+21 more

    mltree-mcp: ML workflow map and experiment tree

    1

    Teach your agent when and how to use the mltree-mcp server: map ML / QSAR / GNN pipelines as a node DAG, log every run as a branching tree, and open or export a KNIME-style map.

    Free

    Security scanned
    mltree-mcp: ML workflow map and experiment tree

    mltree-mcp: ML workflow map and experiment tree

    Example session with this skill installed

    Scan this existing QSAR project with mltree, fix anything the scan got wrong, train the model, and show me the map.

    • Read your context and instructions
    • Compiled the mltree-mcp

    The agent checks the mltree MCP server is connected, calls get_pipeline, previews with scan_project(dry_run=true), writes the map, corrects edges with connect_steps/add_step, runs or logs the train step, calls open_map, and returns http://localhost:8765.

    Connects securely to your tools. The creator never sees your data.

    About this skill

    Teach your agent when and how to use the mltree-mcp server: map ML / QSAR / GNN pipelines as a node DAG, log every run as a branching tree, and open or export a KNIME-style map.

    mltree-mcp is a free, MIT-licensed local MCP server (pip install mltree-mcp or uvx mltree-mcp, Python 3.11+). This skill tells your agent when to scan a project, add and connect pipeline steps, log and compare runs, track sweeps, and open the interactive map at http://localhost:8765. Works with Claude Code, Cursor, Codex and any stdio MCP client. Source: https://github.com/yboulaamane/mltree-mcp

    Requires separate installation of mltree-mcp server. Static scan does not catch dynamic dependencies. Execution features depend on server-side MLTREE_ALLOW_RUN settings.

    How to use

    1. 1

      Connect your agent to the Agensi MCP

      Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup

    2. 2

      Sign in to Agensi

      Your agent shows a sign-in link. Sign in and click Allow.

    3. 3

      Ask for what you need

      Your agent loads this skill and uses it on your task. It always asks before any payment.

    Reviews

    No reviews yet

    Be one of the first to try it. Every listed skill goes through the trust checks below.

    Security scanned

    Checked by our 8-point scan before listing

    Fresh listing

    Recently published to Agensi

    Free forever

    No account required to browse

    Trust & safety

    Security scan

    Security scan passed 5 days ago

    • Free with an Agensi account

    Requires

    Python
    uv (Python tool runner)
    Git
    pip
    uv (Python tool runner)
    run

    Permissions

    Read Files
    Write Files
    Network Access
    Terminal / Shell
    Listed5 days ago

    What's inside

    Frequently Asked Questions

    No. Install the free MIT-licensed server from PyPI or GitHub and connect it to your MCP client. This skill tells the agent when and how to use its 21 tools.

    Any stdio MCP client: Claude Code, Codex, Cursor and others.

    Unlike manual tracking, this skill uses schema-validated tools to automatically capture script snapshots, metrics, and parent-child relationships between trials, ensuring your experiment history remains consistent and reproducible without manual logging.

    You can easily visualize your entire project history using the built-in local web viewer or by generating portable HTML files that contain the full DAG and experiment logs for peer review.

    Yes, the skill includes tools for scanning existing directories to perform static analysis, allowing you to visualize data dependencies and script flows in legacy projects without rewriting your code.

    New in Research & Analysis