experiment

experiment is a skill for Claude Code, Codex from hdu-ailab/EasyResearch. It costs 171 tokens per session (4,677 once invoked), scanned A, original, MIT.

A workspace and process for running reproducible machine-learning and AI experiments for research papers. It organises baselines, datasets, controlled trials, multiple random seeds, ablation studies, and recorded results.

In plain words
What is it for?
Use it to build and compare baselines, test model changes, run local or remote experiments, evaluate datasets, perform ablations, and promote verified results.
Why use it?
It helps prevent unreliable comparisons caused by changing too many things at once or failing to record how a result was produced. It also keeps experiment files in a defined workspace structure.

Skill for Claude CodeCodex

Written for no agent in particular: nothing here depends on one.

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add skills/hdu-ailab/easyresearch/experiment
Any agent
npx skills add hdu-ailab/EasyResearch --skill experiment
Clone the repo
git clone --depth 1 https://github.com/hdu-ailab/EasyResearch

Made for: Claude Code, Codex.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for experiment

README.md
[![agentmods](https://agentmods.dev/badge/skills/hdu-ailab/easyresearch/experiment.svg)](https://agentmods.dev/skills/hdu-ailab/easyresearch/experiment)
Your own site
<a href="https://agentmods.dev/skills/hdu-ailab/easyresearch/experiment"><img src="https://agentmods.dev/badge/skills/hdu-ailab/easyresearch/experiment.svg" alt="Measured on agentmods" height="20"></a>
Per session 171 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 4,677 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5.1 $0.00171 $0.04677
Opus 5 $0.00086 $0.02338
Sonnet 5 $0.00034 $0.00935
Haiku 4.5 $0.00017 $0.00468

Measured 6d ago against content hash 55c8b9df833c, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-06, from the pricing page.

Security

Grade A, and why

experiment scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 6d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

src/skills/experiment/SKILL.md · 406 lines

How it starts

The opening of the file, as written. The whole thing — 406 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Experiment Workspace

Purpose

Use this skill to constrain research experiment workspaces and enforce a paper-driven experimental process. The goal is to find a model that is either clearly stronger than baselines or more innovative while remaining experimentally defensible.

Most practical model innovations are not invented from nothing. They are discovered by reading papers, identifying what each component is useful for, systematically recombining components from different papers, and validating which combinations actually work.

Do not jump directly to a proposed model. Read papers first, build baselines first, then try component combinations.

In the bundled paper workflow, select exactly one root from the dispatch before editing: exact-cwd experiments/ for local execution or the Research Assistant-created and marker-verified exact-cwd experiment_ssh/ mount for SSH execution. Do not mix or mirror the two roots.

Required Directory Layout

Create or maintain this structure under the user-specified experiment root:

<experiment-root>/
  .venv/
  src/
  external/
    baselines/
      <baseline_name>/
        repo/
        .venv/
    benchmarks/
      <benchmark_name>/
        repo/
        .venv/
  outputs/
  results/
  logs/
  datasets/
  experiment-record.md

For local mode, <experiment-root> is experiments/. For SSH mode, it is only experiment_ssh/, which is the local view of the configured remote project root. An explicitly supplied existing user layout may replace the local-mode root, but it never replaces the verified bundled SSH mount.

Directory rules:

  • .venv/: main experiment environment managed by uv; use it only for this project's own experiment code, routers, models, analysis, plotting, and lightweight adapters.
  • src/: source code written for this project, including data loading, models, training, evaluation, adapters, analysis, and utilities.
  • external/: all external author repositories, official benchmark repositories, and their isolated environments. Do not place cloned third-party repos directly under src/.
  • external/baselines/<baseline_name>/repo/: checkout or symlink to the author-maintained baseline repository.
  • external/baselines/<baseline_name>/.venv/: isolated uv environment for that baseline only.
  • external/benchmarks/<benchmark_name>/repo/: checkout or symlink to the official benchmark/evaluator repository.
  • external/benchmarks/<benchmark_name>/.venv/: isolated uv environment for that benchmark only.
  • outputs/: all raw outputs from experiment scripts, including failed trials, temporary models, trial models, and formal model runs.
  • results/: clean copy of formal experiment outputs only. Use this directory later for paper writing, figures, statistical analysis, and tables.
  • logs/: command logs, runtime logs, error traces, and environment snapshots.
  • datasets/: raw or processed datasets, plus dataset manifests when useful.
  • experiment-record.md: human-readable experiment progress record. Update it after every experiment run.

Read the full file on GitHub · 406 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 6d ago First seen · 406 lines · 171 tokens per session scan A 55c8b9df833c

Subscribe to this mod's changes

experiment is a skill published in the GitHub repository hdu-ailab/EasyResearch (11 stars, last pushed 2d ago), licensed MIT. It adds 171 tokens to every session and 4,677 once invoked, about $0.0009 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.

Related

Other skills, from other repositories

alterlab-pyhealth

Develops, tests, and deploys clinical machine learning models with the PyHealth healthcare AI toolkit. Use when working with electronic health records (EHR), clinical prediction tasks (mortality, readmission, drug recommendation), medical coding systems (ICD, NDC, ATC), physiological signals (EEG, ECG), healthcare…

AlterLab-IEU/AlterLab-Academic-Skills · 117 tokens

alterlab-deepchem

Runs molecular machine learning with DeepChem — diverse featurizers, pre-built MoleculeNet benchmark datasets, and pre-trained models (ChemBERTa, GROVER) for property prediction (ADMET, toxicity, solubility) via traditional ML or graph neural networks. Use when running end-to-end molecular ML experiments that need…

AlterLab-IEU/AlterLab-Academic-Skills · 126 tokens

alterlab-esm

Run ESM protein language models — ESM3 for generative multimodal protein design across sequence, structure, and function, and ESM C for efficient embeddings and representations — locally or via the cloud Forge API. Use when working with protein sequences, structures, or function prediction, designing novel proteins…

AlterLab-IEU/AlterLab-Academic-Skills · 89 tokens

alterlab-molfeat

Featurizes molecules for machine learning with molfeat (100+ featurizers) — ECFP/MACCS/MAP4 fingerprints, RDKit and Mordred physicochemical descriptors, and pretrained embeddings (ChemBERTa, ChemGPT, GIN) exposed as scikit-learn transformers that convert SMILES into feature vectors. Use when turning molecules into…

AlterLab-IEU/AlterLab-Academic-Skills · 143 tokens

alterlab-geniml

Machine learning on genomic interval data (BED files) with the geniml Python package — region embeddings (Region2Vec), joint region+metadata embeddings (BEDspace/StarSpace), single-cell ATAC-seq embeddings (scEmbed), consensus peak sets / universes (build-universe), tokenization, BEDshift randomization, and…

AlterLab-IEU/AlterLab-Academic-Skills · 152 tokens

alterlab-gtars

Runs high-performance genomic interval analysis with gtars (databio), a Rust toolkit with Python bindings — the performance-critical backend for the geniml ML library. Use when computing overlaps/jaccard/coverage between BED region sets, indexing intervals with IGD, generating uniwig accumulation/coverage tracks…

AlterLab-IEU/AlterLab-Academic-Skills · 132 tokens