# Executable Article

> >

- **Type:** Skill
- **Install:** `agentstack add skill-neuroanalytics-data-science-harness-executable-article`
- **Verified:** Yes — security-reviewed for prompt injection and unsafe behavior
- **Seller:** [neuroanalytics](https://agentstack.voostack.com/s/neuroanalytics)
- **Installs:** 0
- **Category:** [Agent Skills](https://agentstack.voostack.com/c/agent-skills)
- **Latest version:** 0.1.0
- **License:** MIT
- **Upstream author:** [neuroanalytics](https://github.com/neuroanalytics)
- **Source:** https://github.com/neuroanalytics/data-science-harness/tree/main/plugins/disseminate/skills/executable-article

## Install

```sh
agentstack add skill-neuroanalytics-data-science-harness-executable-article
```

Requires the [AgentStack CLI](https://agentstack.voostack.com/docs/cli). Works with Claude Code, Cursor, and any MCP-compatible agent.

## About

# Skill: executable-article

Turn a product into a **re-executable article** — figures regenerate from the provenanced outputs in
the project's pinned container, rather than being pasted in. That makes the paper Actionable
(re-runs) and Portable/Ephemeral (rebuilt from spec). You delegate history/ledger reads and the save
to the **datalad doer**.

Load `plugins/disseminate/references/neurolibre-structure.md` for the scaffold and how each piece
maps to the harness before generating.

## When to use
- A product (usually a paper) is analyzed and (ideally) released, and the author wants a
  reproducible executable-article form.
- Do NOT use to write prose (`draft-manuscript`) or to release/DOI a version (`dataset-release`);
  this generates the executable build around the results.

## Steps
1. **Identify inputs** — the product `id`, its comparisons' outputs (`derivatives/cmp-/`), the
   `containers/` recipe (for the environment), and the dataset's published location (the
   `dataset-release` DOI or the DataLad sibling — from the ledger). If the dataset is unreleased,
   note that repo2data should point at the sibling and a DOI be wired at release.
2. **Scaffold the article** (per the reference) at `article/` (or the author's path): `myst.yml`,
   `paper.md`, `content/` figure notebooks, and `binder/` with an environment derived from
   `containers/` plus a `data_requirement.json` (repo2data) resolving the published dataset.
3. **Wire figures to provenance** — each `content/` notebook computes its figure from the
   provenanced `derivatives/…` outputs (not a static image), so the build reproduces them.
4. **Register + log** — add the article path to the product's `outputs[]`; append
   `{ ts, op: executable-article, stage: disseminate, note: "NeuroLibre article for ", branch:  }`.
5. **Save** — delegate to the datalad doer: "save: `datalad save -m 'executable-article: scaffold '`."
6. **Report** — the article path, how to build/preview (`myst build` / Jupyter Book), what still
   needs wiring (e.g. the dataset DOI once released), and the next step: `link-outputs` to relate the
   article to the dataset (`Documents`) and paper (`IsSupplementTo`).

## Constraints
- Figures must regenerate from provenanced outputs + the pinned environment — do not embed static
  images as the source of truth; the article's value is re-executability.
- The `binder/` environment derives from the project's container recipe/digest — keep it consistent
  with what analyses actually ran in; do not invent dependencies.
- repo2data points at the *published* dataset (DOI or sibling) — never a local absolute path.
- Record the article under the product's `outputs[]`; keep `log:` append-only and the ledger
  schema-valid. Delegate reads/saves to the datalad doer.

## Source & license

This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.

- **Author:** [neuroanalytics](https://github.com/neuroanalytics)
- **Source:** [neuroanalytics/data-science-harness](https://github.com/neuroanalytics/data-science-harness)
- **License:** MIT

Install and usage instructions live in the source repository linked above.

## Pricing

- **Free** — Free

## Security capabilities

Automated source analysis of v0.1.0 — what this tool can access:

- **Network access:** no
- **Filesystem access:** no
- **Shell / process execution:** no
- **Environment & secrets:** no
- **Dynamic code execution:** no

*"Yes" means the capability is present in the source — more access means more to trust, not that it is unsafe.*


## Versions

- **0.1.0** — security scan: passed — Imported from the upstream source.

## Links

- Listing page: https://agentstack.voostack.com/l/skill-neuroanalytics-data-science-harness-executable-article
- Seller: https://agentstack.voostack.com/s/neuroanalytics
- Browse the marketplace: https://agentstack.voostack.com/browse

---
Listed on AgentStack — the marketplace for AI agent skills and MCP servers. Every listing is security-reviewed. Creators keep 70%.