More screenshots

    Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIMuseMuseDotsDots+21 more

    experiment reproducibility auditor

    1

    Audit and verify the reproducibility of ML experiments by mapping code, data, and environment lineage.

    $9

    Secure checkout via Stripe

    Money-back guarantee

    Converts to your local currency at checkout

    Security scanned
    experiment reproducibility auditor

    experiment reproducibility auditor

    Example session with this skill installed

    Audit this MLflow run (ID: 5f2b8) against Git commit a1b2c3d. Here are the DVC data hashes for the training set. Can we reproduce the accuracy result of 0.94?

    • Read your context and instructions
    • Compiled the experiment reproducibility auditor

    ARTIFACT: EXPERIMENT-REPRO-AUDIT
    VERDICT: Incomplete lineage.

    Findings

    • Code: [Verified] Matches commit a1b2c3d.
    • Data: [Verified] DVC hashes match.
    • Environment: [Assumed] Python 3.9 (missing manifest).
      Implementation: Re-export conda.yaml from run 5f2b8 to verify environment consistency.
    Re-export conda.yaml

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

    About this skill

    The problem

    Machine learning experiments often fail to replicate because of missing lineage between code, data, and environment. Teams struggle to verify if a specific model artifact actually matches a specific MLflow run or Git commit.

    What it does

    • Maps Git commits, MLflow metadata, and DVC hashes into a single evidence chain.
    • Identifies missing links in the reproduction pipeline before scoring results.
    • Flags statements as Verified, Inferred, or Assumed to prevent false confidence.
    • Generates a structured reproducibility audit with implementation steps for recovery.

    Frameworks & tools

    MLflow, DVC, Git, and various experiment manifests or logs.

    Why this beats prompting it yourself

    Standard LLM prompts often assume tool presence implies configuration maturity. This skill enforces a strict evidence contract that refuses to infer capability from tool names and explicitly tracks unverified assumptions through a specialized suite trace.

    Use cases

    • Auditing a model run to ensure it can be rebuilt from scratch.
    • Identifying why a local experiment doesn't match a remote baseline.
    • Verifying lineage for regulatory compliance or peer review.
    • Mapping disparate metadata from multiple ML tools into one source of truth.

    Known limitations

    Does not debug failed training runs or select orchestrators. Does not execute deployments or handle credentials.

    Does not debug code errors in failed training runs or manage infrastructure/deployments. Requires manual input of metadata from tools like MLflow or DVC.

    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

    Money-back guarantee

    Doesn't do what its page promises? You get your money back.

    Trust & safety

    Security scan

    Security scan passed 2 months ago

    Permissions

    No special permissions declared or detected

    Listed2 months ago

    What's inside

    Frequently Asked Questions

    The skill acts as a forensic validator that links Git commits, DVC data hashes, and MLflow metadata to ensure a model can be perfectly recreated, preventing the "it works on my machine" syndrome in production environments.

    It is designed to work with any AI agent or orchestration platform that supports MLflow and DVC integration, requiring only read-access to your experiment manifests and repository logs.

    You receive the pre-configured skill manifest, the logic for the multi-tool evidence chain, and a structured template for generating audit reports that categorize lineage as Verified, Inferred, or Assumed.

    The auditor requires access to your experiment metadata (such as MLflow run IDs or DVC lockfiles) but does not need to execute your training code or handle sensitive cloud provider credentials.

    While it maps the lineage of existing experiments, it does not debug code errors that cause training to fail or suggest which orchestrator you should use for deployment.

    New in Agents & Orchestration