M

ML Pipeline

Enforces a strict 16-step data-first ML workflow — data inspection, EDA, and pro

jananthan-paramsothy/ml-pipeline · v0.4.0 · Development & Workflow

by Jananthan ParamsothyOwner verifiedAwesome AI Plugins

Ask a coding agent to "train a model on this data" and it calls .fit() within five minutes. Often it fits the scaler before splitting — and that leak doesn't surface until production. It isn't being careless. It's being helpful the way it was trained to be: a number, fast. The number is wrong in a way that looks fine. ml-pipeline makes that impossible. Sixteen steps, fixed order, data first. The split happens before feature engineering. The test set is touched once, at the end. At four gates the agent stops, shows its work with figures, and waits for your go-ahead. • marimo notebooks, matplotlib figures at every data step • PIPELINE.md logs every step and approval — sessions resume, not restart • Security scan: 90/100 (A), zero critical or high findings • Apache-2.0, one 7.8 KB skill file, no dependencies • Works in Claude Code, Codex, and Kimi Code

Follow1
Protect this install with HOL Guard

Guard checks the skills and MCP servers a plugin adds on your machine and asks you before a risky action runs. The scores above are free, and no plan changes them.

Is ML Pipeline a fit?

From the plugin's manifest and HOL's scan. Anything the plugin does not declare is shown as not declared.

Works with

  • Claude Code
  • Kimi Code
  • Codex
  • 1 skill

Before you install

Sign-in
Asks you to sign in when you install it
Declared access
Not declared
License
Apache-2.0

Check before relying on it

What is ML Pipeline?

ML Pipeline is a published development & workflow plugin for AI coding agents that works with Claude Code, Kimi Code and Codex, developed by Jananthan Paramsothy and distributed through the HOL AI plugin registry. Enforces a strict 16-step data-first ML workflow — data inspection, EDA, and problem definition before any model training — with explicit user-permission gates, marimo notebooks, and matplotlib figures at every step; also works in Claude Code and Kimi.

Canonical slug
jananthan-paramsothy/ml-pipeline
Version
v0.4.0 · updated Nov 2, 2026

Registry snapshot

Owner verification
Verified on HOL as @jananthan30
Marketplace source
.claude-plugin/marketplace.json
Scanner
HOL plugin scanner
Security result
Partial scan
Digest verified
Yes
1 bundled skillRead, copy or download each SKILL.mdOpen skills

Trust and security

Trust & reputation

On-Chain Verified
92

HOL Trust Score

Per-metric points (0–100 each) combined via a weighted average into the overall score.

Installability
100/100
Maintenance
65/100
MCP server setup
100/100
Plugin Security
95/100
Provenance
80/100
Publisher Quality
100/100

Provenance

Plugin root
.
Source repo
https://github.com/jananthan30/ml-pipeline
Source commit
a47fa7475ca3…
Publisher verified
Yes
Owner verified
@jananthan30
Agent Plugins 1.0
Not detected

Continuous scanner CI detected

No action is required. This plugin receives the full trust score.

README badge not detected

Plugin owners can add the HOL badge to the repository README from Guard Plugins. Distribution, not security evidence. It adds a 2% engagement adjustment to the overall trust score and never changes the security result.

Partial scan
Security result
95
Score from finished checks
0
High findings

Some checks did not finish: Cisco skill scan. No blocking issue was found in the checks that did finish, but this is not a passed scan.

Provider
codex-plugin-scanner
Grade
A · Excellent
Version
3.24.2
codex:hol-plugin-scanner-qFZj0Q / cisco-skill-scanner: failedcodex:hol-plugin-scanner-qFZj0Q / cisco-mcp-scanner: skipped

[claude:.] Claude Plugin

18/18

5 checks

[claude:.] Security

16/16

6 checks

[claude:.] Operational Security

N/A

5 checks · not applicable

[claude:.] Code Quality

10/10

2 checks

[codex:.] Manifest Validation

28/31

9 checks

[codex:.] Security

16/16

9 checks

[codex:.] Operational Security

N/A

5 checks · not applicable

[codex:.] Best Practices

9/12

5 checks

[codex:.] Marketplace

15/15

3 checks

[codex:.] Skill Security

5/5

4 checks

[codex:.] Code Quality

10/10

2 checks

Published advisories

None matched in GitHub security advisories, checked . This covers published advisories only, not problems nobody has reported.

GitHub security advisories are those the source repository has published.

Findings

infomanifest-validationPLUGIN_JSON_INTERFACE_ASSET_PRIVACYPOLICYURL

The interface field "privacyPolicyURL" must use HTTPS or point to a safe in-repo asset.

Update "privacyPolicyURL" to use HTTPS or an existing relative asset path.

infomanifest-validationPLUGIN_JSON_INTERFACE_ASSET_TERMSOFSERVICEURL

The interface field "termsOfServiceURL" must use HTTPS or point to a safe in-repo asset.

Update "termsOfServiceURL" to use HTTPS or an existing relative asset path.

infomanifest-validationPLUGIN_JSON_INTERFACE_ASSET_COMPOSERICON

The interface field "composerIcon" must use HTTPS or point to a safe in-repo asset.

Update "composerIcon" to use HTTPS or an existing relative asset path.

infomanifest-validationPLUGIN_JSON_INTERFACE_ASSET_LOGO

The interface field "logo" must use HTTPS or point to a safe in-repo asset.

Update "logo" to use HTTPS or an existing relative asset path.

infomanifest-validationPLUGIN_JSON_INTERFACE_ASSET_SCREENSHOTS

The interface field "screenshots" must use HTTPS or point to a safe in-repo asset.

Update "screenshots" to use HTTPS or an existing relative asset path.

infobest-practicesCODEXIGNORE_MISSING

A .codexignore file helps prevent accidental inclusion of local artifacts and secrets.

Add a .codexignore file with generated assets, local state, and secret paths.

Install guidance

Install in Claude Code

Install through the Claude Code plugin marketplace.

Claude Code plugin docs

Optional: install HOL Guard for Claude Code first, so it can check this plugin's tools on your machine before Claude Code uses them.

  1. 1

    Add the marketplace

    Run this inside a Claude Code session. It registers the marketplace manifest shipped at the repository root. If the repository ships none, add the generated entry from the Advanced section below first.

    claude code
  2. 2

    Install the plugin

    Installs the ml-pipeline entry from the ml-pipeline marketplace.

    claude code
  3. 3

    Scripted alternative

    Non-interactive equivalent for scripts and CI pipelines. Add --scope project to pin the install to one repository.

    shell
  4. 4

    Check it is installed

    Lists installed plugins. The entry should appear as enabled.

    shell

To remove it later, run claude plugin uninstall ml-pipeline@ml-pipeline.

Did setup work in Claude Code?

No setup reports for Claude Code yet. Self-reported by signed-in visitors. Copying a command is not counted, and most installs are never reported.

Attested installs: Fewer than 10 attested installs of this plugin in week 2026-W46, so no number is shown. Counted from opt-in HOL Guard install reports, once per machine and runtime, and never mixed with the answers above. How install reports work.

1 cover
1 / 4

1 cover

Full size
Loading reviews

Plugin manifest

{
  "name": "ml-pipeline",
  "version": "0.4.0",
  "description": "Data-first ML discipline: forces coding agents through a strict 16-step pipeline — inspect and understand the raw data, define the prediction problem, clean, engineer, and split it BEFORE any model training — with explicit user-permission gates, marimo notebooks, and matplotlib visuals at every step.",
  "author": {
    "name": "Jananthan Paramsothy"
  },
  "license": "Apache-2.0",
  "homepage": "https://github.com/jananthan30/ml-pipeline",
  "repository": "https://github.com/jananthan30/ml-pipeline",
  "keywords": [
    "machine-learning",
    "ml-pipeline",
    "eda",
    "data-cleaning",
    "data-leakage",
    "marimo",
    "matplotlib",
    "workflow-discipline"
  ],
  "skills": "./",
  "interface": {
    "displayName": "ML Pipeline",
    "developerName": "Jananthan Paramsothy",
    "category": "Productivity",
    "shortDescription": "Stops coding agents from jumping straight to model training.",
    "longDescription": "Enforces a strict 16-step, data-first ML pipeline: data inspection, EDA, and prediction-problem definition before any cleaning or modeling; train/val/test split before feature engineering and preprocessing; a baseline before any complex model; and a final test set that is touched exactly once. Explicit user-permission gates after each phase, marimo notebooks as the workbench, matplotlib figures at every data-facing step, and a per-project PIPELINE.md checklist so sessions resume instead of restarting.",
    "description": "Stops coding agents from jumping straight to model training. 16 strict steps, 4 permission gates, visuals at every step.",
    "capabilities": [
      "skills"
    ],
    "websiteURL": "https://github.com/jananthan30/ml-pipeline"
  },
  "holManifestOrigin": "repository",
  "registryIndexVersion": 9
}

Marketplace source

Repo URL
https://github.com/jananthan30/ml-pipeline
Marketplace path
.claude-plugin/marketplace.json
Source path
plugins/jananthan30/ml-pipeline
Install policy
AVAILABLE

Skills

1 SKILL.md file ship with this plugin. Preview, copy, or download each one, then install them with the Skills CLI.

Share
skills-cli
  • ml-pipeline

    skills/ml-pipeline/SKILL.md

    MANDATORY whenever a task involves training, fine-tuning, tuning, or evaluating a machine-learning model on data (tabular, time series, text, images — any modality). Enforces a strict 16-step pipeline that starts with inspecting the raw data, gates each phase behind the user's explicit permission, and produces marimo notebooks with matplotlib visuals so the user can see and understand every step. Never jump straight to model training.

    SKILL.md content is not in this snapshot. Open the raw file to copy it from the source repository.

File inventory

.agents/plugins/marketplace.json

file

355 bytes

016b1ff16b286228…

.claude-plugin/marketplace.json

file

651 bytes

548a1b205a5e8c1e…

.claude-plugin/plugin.json

file

631 bytes

effedf94b1d976af…

.codex-plugin/plugin.json

plugin-manifest

1,676 bytes

491dc4b0f7a507a3…

hooks/guard_training.py

file

19,229 bytes

dc8b7fed4f660998…

hooks/hooks.json

file

1,042 bytes

70b2e178a635ac6d…

hooks/post_tool_warn.py

file

4,020 bytes

c527a53ea6807d1e…

hooks/session_start.py

file

1,283 bytes

83c72ead365b63cc…

LICENSE

file

15,144 bytes

01cd5191f2c0ccdb…

README.md

file

3,801 bytes

3c8583d533acfdfd…

SECURITY.md

file

813 bytes

8fdd751ebe3f5f4b…

skills/ml-pipeline/references/model-selection.md

skill

5,064 bytes

3b3dd2b87a67521c…

skills/ml-pipeline/SKILL.md

skill

13,255 bytes

07dd4658ecbd577e…

ML Pipeline — Frequently asked questions

What is ML Pipeline?
ML Pipeline is an AI plugin in the HOL registry. Ask a coding agent to "train a model on this data" and it calls .fit() within five minutes. Often it fits the scaler before splitting — and that leak doesn't surface until production. It isn't being careless. It's being helpful the way it was trained to be: a number,…
How do I install ML Pipeline?
Install ML Pipeline in your harness: Codex — codex plugin marketplace add jananthan30/ml-pipeline; Claude Code — /plugin marketplace add jananthan30/ml-pipeline; Any agent — npx [email protected] add jananthan30/ml-pipeline. Full step-by-step guidance is on the HOL plugin page.
How do I install ML Pipeline in HOL Guard?
To install ML Pipeline in HOL Guard, start with hol-guard plugin install jananthan-paramsothy/ml-pipeline --apply. The complete step-by-step install guide for HOL Guard is on the HOL plugin page.
How do I install ML Pipeline in Codex?
To install ML Pipeline in Codex, start with codex plugin marketplace add jananthan30/ml-pipeline. The complete step-by-step install guide for Codex is on the HOL plugin page.
How do I install ML Pipeline in Claude Code?
To install ML Pipeline in Claude Code, start with /plugin marketplace add jananthan30/ml-pipeline. The complete step-by-step install guide for Claude Code is on the HOL plugin page.
Is ML Pipeline free?
ML Pipeline is listed as free on HOL.
How is the Trust Score for ML Pipeline computed?
ML Pipeline has a HOL Trust Score of 92 out of 100. The score is a weighted sum of six components from a static scan of the source, capped for high-severity findings and adjusted by two disclosed policies. Memberships and payments are not inputs, and a scan is not a safety guarantee. Full breakdown: https://hol.org/registry/plugins/jananthan-paramsothy%2Fml-pipeline/trust
Who publishes ML Pipeline?
ML Pipeline is published by Jananthan Paramsothy and listed on HOL.
Is ML Pipeline available now?
ML Pipeline is currently available according to its latest public launch.

Directory acceptance

Listed in 1 of 5 directory lists

Read from the list each directory publishes on GitHub. Being listed means the directory accepted the plugin; it is not a HOL security rating.

Lists last read . HOL re-reads them every day.

Submitting it? Check it against each directory's rules

Install signal

HOL cannot see every install. These are separate signals from different sources, so they are not added together. Counts under each floor show as <N.

Install commands issued here
<5
HOL Guard install commands issued from this listing, last 28 days.
Opt-in install reports
<10
Machines that reported installing it through HOL Guard in week 2026-W46. Reporting is opt-in, so most installs are not counted.
HOL Guard workspaces
<5
Workspaces that approved it or include it in an approved-plugin feed, as of today.
New GitHub stars
<5
From HOL's GitHub snapshots of the repository, 2026-10-25 to 2026-11-22.

History, reviews, and similar plugins

Latest launch