inlab-human

v0.2.1

In-lab human interaction with deployed prediction endpoints — the Claude Code chat is the UI. CONSOLE mode (default): patient-first on-demand inference — list patients, view the full chart, list available models (Endpoint_Set packages + live status), and run 'this model on this patient': the endpoint-predict MCP tool prepares the payload (required tables + trigger record), POSTs to the endpoint, and returns the score verbatim with a data-gaps report. STUDY mode: formal reader protocol (frozen review_bundle.json, blind→assisted session, decision-influence metrics vs gold). Endpoint-agnostic: local Flask, Databricks, SageMaker — same wire contract.

Claude Code5 Skills

By Junjie LuoLicense: MIT1 GitHub starsUpdated last week

Directory evidence

Runtimes: Claude Code
Parsed components: 5 skill or MCP entries
Source updated: Sep 16, 2026
Manifest status: Canonical path parsed

The directory validates manifest shape and source location. It does not execute the plugin or provide a security endorsement. Review the indexing methodology →

Install inlab-human for Claude Code

Installs for the current user

claude plugin marketplace add IchenDEV/agent-plugin-mkt
claude plugin marketplace update agent-plugin-marketplace
claude plugin install inlab-human@agent-plugin-marketplace

Paste and run these commands in a terminal with Claude Code. They add and refresh the PluginsMP catalog, then install this plugin.

The installer fetches third-party code from the source repository shown on this page. This directory validates manifest structure and source location, but does not perform a security audit; review the manifest, components, and source before installing.

Get the source manually

$ git clone https://github.com/jluo41/Tools

Clone the source repository, then follow its setup instructions to add the plugin to a compatible client. The plugin root is plugins/inlab-human/.

Plugin files

plugins/inlab-human/

├── .claude-plugin/plugin.json
├── skills/inlab-human/SKILL.md
├── skills/inlab-human-bundle/SKILL.md
├── skills/inlab-human-console/SKILL.md
├── skills/inlab-human-report/SKILL.md
└── skills/inlab-human-review/SKILL.md

Included Skills5

inlab-humanskills/inlab-human/SKILL.md

In-lab human interaction with deployed prediction endpoints — the Claude Code chat is the UI. Tier-1 orchestrator with two modes: CONSOLE (default; patient-first on-demand inference: pick patient → see chart → list models → predict via the endpoint-predict tool → analyze) and STUDY (formal reader protocol: bundle → blind/assisted review → decision-influence report). Endpoint-agnostic: local Flask, Databricks, SageMaker — same wire contract. Trigger: inlab, console, get patient, run prediction, clinician eval, reader study, /inlab-human.

inlab-human-bundleskills/inlab-human-bundle/SKILL.md

Freeze a review_bundle.json for an in-lab reading session: sample de-identified cases, score each via the endpoint-predict tool (live endpoint, deterministic), add narrator-agent explanations, validate schema invariants, freeze. v0.1 source mode: the Endpoint_Set package's own examples/. Called by /inlab-human orchestrator; also directly. Trigger: build bundle, freeze bundle, review bundle, /inlab-human-bundle.

inlab-human-consoleskills/inlab-human-console/SKILL.md

The clinician-driven inference console — the default inlab-human mode. Patient-first flow in the Claude Code chat: pick a patient from the patient store, see their full chart, list the available prediction models (live Endpoint_Set endpoints), run 'this model on this patient' — the endpoint-predict tool prepares the payload, POSTs to the endpoint, returns the score — then analyze/explain the result. On-demand, interactive; no bundles, no blinding. Trigger: console, get patient, list models, run prediction for patient, /inlab-human-console.

inlab-human-reportskills/inlab-human-report/SKILL.md

Score an in-lab reading session: join responses.jsonl with the bundle's gold outcomes and compute model accuracy on the sample, clinician blind-vs-assisted accuracy, decision-influence split by model correctness (healthy reliance vs over-reliance — the safety readout), rating distributions, and time-per-case. Emits metrics.json + a markdown report. Called by /inlab-human orchestrator; also directly. Trigger: inlab report, session metrics, reader study results, /inlab-human-report.

inlab-human-reviewskills/inlab-human-review/SKILL.md

Run the in-lab clinician reading session in the Claude Code chat: blind pass (raw case data only → clinician's risk estimate, decision, confidence), reveal (model score + explanation), assisted pass (re-estimate + rate score and explanation SEPARATELY), append structured rows to responses.jsonl. Consumes a frozen review_bundle.json; never calls a live endpoint; never shows gold. Called by /inlab-human orchestrator; also directly. Trigger: reading session, review bundle with clinician, blind assisted, /inlab-human-review.

Plugin manifests1

plugins/inlab-human/.claude-plugin/plugin.json

{
  "name": "inlab-human",
  "version": "0.2.1",
  "description": "In-lab human interaction with deployed prediction endpoints — the Claude Code chat is the UI. CONSOLE mode (default): patient-first on-demand inference — list patients, view the full chart, list available models (Endpoint_Set packages + live status), and run 'this model on this patient': the endpoint-predict MCP tool prepares the payload (required tables + trigger record), POSTs to the endpoint, and returns the score verbatim with a data-gaps report. STUDY mode: formal reader protocol (frozen review_bundle.json, blind→assisted session, decision-influence metrics vs gold). Endpoint-agnostic: local Flask, Databricks, SageMaker — same wire contract.",
  "author": {
    "name": "Junjie Luo",
    "url": "https://github.com/jluo41"
  },
  "license": "MIT",
  "keywords": [
    "human-evaluation",
    "clinician",
    "reader-study",
    "blind-assisted",
    "risk-prediction",
    "endpoint",
    "shap",
    "decision-influence",
    "ground-truth",
    "human-in-the-loop"
  ]
}

For maintainers

If you maintain this plugin, link to this source-backed listing from your README so users can review its manifest and indexed components.

[inlab-human on Agent Plugins Marketplace](https://pluginsmp.com/plugins/inlab-human)