# Datalad Siblings

> >

- **Type:** Skill
- **Install:** `agentstack add skill-neuroanalytics-data-science-harness-datalad-siblings`
- **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/datalad-cli/skills/datalad-siblings

## Install

```sh
agentstack add skill-neuroanalytics-data-science-harness-datalad-siblings
```

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

## About

# Skill: datalad-siblings

Inspect and configure sibling remotes — named locations where dataset history and/or
annexed file content can be pushed and pulled.

## Steps

1. **Verify DataLad context** — check for `.datalad/` in the current directory or a
   parent:
   ```bash
   ls .datalad/ 2>/dev/null
   ```
   If no dataset is found, inform the user and stop.

2. **Determine action** — read from `$ARGUMENTS` or conversation context:
   - **query** (default, no subcommand): list all configured siblings
   - **add**: register a new sibling
   - **configure**: change a property of an existing sibling
   - **remove**: unregister a sibling
   - **enable**: activate a special remote (e.g., S3, OSF) after cloning

3. **Execute based on action**:

   ### query
   ```bash
   datalad siblings
   ```
   Present the output. Explain the `+`/`-` indicator and sibling types.

   ### add
   Gather from the user:
   - Sibling name (e.g., `github`, `osf-storage`, `backup`)
   - URL (fetch URL — see reference for format options)
   - Whether a separate `--pushurl` is needed (common when fetch is HTTPS but push
     requires SSH):
     > "Do you need a separate push URL (e.g., SSH for write access)?"
   - Whether a `--publish-depends` should be set — ask if a storage sibling already
     exists or is being added alongside a git host:
     > "Should pushing to this sibling automatically push annexed content to a storage
     > sibling first? If so, which storage sibling?"

   Construct:
   ```
   datalad siblings add -s  \
       --url  \
       [--pushurl ] \
       [--publish-depends ]
   ```

   ### configure
   Gather which sibling and which property to change. Common properties:
   - `--url` / `--pushurl`
   - `--annex-wanted` / `--annex-required`
   - `--publish-depends`

   ```
   datalad siblings configure -s  [--annex-wanted '']
   ```

   ### remove
   Confirm the sibling name. Warn that this only removes the registration — it does
   not delete remote content.
   ```
   datalad siblings remove -s 
   ```

   ### enable
   Used after cloning a dataset whose special remote (OSF, S3, WebDAV) needs activation:
   ```
   datalad siblings enable -s 
   ```

   ### create-sibling-github / create-sibling-gitlab / create-sibling-ria / create-sibling-gin
   When the user wants to create a new remote repository and register it as a sibling in
   one step, use the appropriate `create-sibling-*` command. Load
   `${CLAUDE_PLUGIN_ROOT}/../references/siblings-and-remotes.md` for platform-specific
   flags, then:

   a. Identify the target platform (GitHub, GitLab, RIA, GIN, WebDAV, or generic SSH).
   b. Gather required parameters (repo name, organization/namespace, access token if
      needed, RIA store path if applicable).
   c. Construct and display the command before executing:

   **GitHub:**
   ```bash
   datalad create-sibling-github \
     --dataset . \
     --reponame  \
     [--github-organization ] \
     [--access {read|write}] \
     [-s github] \
     [--publish-depends ]
   ```
   Use `--access read` to create a read-only repository (default: `write`).

   **GitLab:**
   ```bash
   datalad create-sibling-gitlab \
     --dataset . \
     --reponame / \
     --gitlab-host  \
     [--access {read|write}] \
     [-s gitlab] \
     [--publish-depends ]
   ```
   Use `--access read` to create a read-only repository (default: `write`).

   **RIA store:**
   ```bash
   datalad create-sibling-ria \
     --dataset . \
     --name ria-storage \
     ria+ssh://user@host/path/to/ria-store
   ```

   **GIN:**
   ```bash
   datalad create-sibling-gin \
     --dataset . \
     --reponame  \
     [-s gin]
   ```

   **WebDAV:**
   ```bash
   datalad create-sibling-webdav \
     --dataset . \
     --url webdavs:///path/to/dataset \
     [-s webdav]
   ```

   **OSF (Open Science Framework):**
   ```bash
   datalad create-sibling-osf \
     --dataset . \
     --title "" \
     [-s osf-storage]
   ```
   Note: requires the `datalad-osf` extension (`pip install datalad-osf`). Verify it is
   installed before suggesting this option.

   d. After creation, suggest pushing: `datalad push --to `.

   If the user asks for `create-sibling` (generic SSH/local path), gather the URL and
   construct: `datalad create-sibling --name  --url `.

4. **Show command and execute** — always display the full command before running.

5. **Post-add suggestion** — after adding or configuring a sibling, suggest:
   > "Run `datalad push --to ` to test connectivity and push your current state."

## Reference

Always load `${CLAUDE_PLUGIN_ROOT}/../references/siblings-and-remotes.md` for URL
formats, special remote types, platform flags, publish-depends patterns, and
annex-wanted expressions.

## Constraints

- Never conflate a **sibling** (DataLad concept, wraps git remote + annex) with a raw
  `git remote` — direct `git remote add` bypasses annex configuration.
- Never skip the `--publish-depends` question when a storage sibling exists alongside a
  git host — failing to set it causes `datalad push` to push git history without annexed
  content, breaking reproducibility for downstream consumers.
- Always show the full command before executing.
- For `remove`, always confirm the sibling name before proceeding — it cannot be undone
  without re-adding.

## 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-datalad-siblings
- 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%.