figshare-skill

name	figshare-skill
description	Use whenever the user wants to interact with Figshare - searching public datasets/articles, downloading Figshare files, listing their own articles/collections/projects, creating or updating articles, or uploading files (including large multi-part uploads) via the Figshare v2 REST API. Trigger on mentions of "figshare", figshare DOIs (10.6084/m9.figshare.*), figshare.com URLs, or phrases like "upload my dataset to figshare", "publish to figshare", "get figshare article".
homepage	https://github.com/Agents365-ai/figshare-skill
license	MIT

Figshare Skill

Interact with the Figshare v2 REST API to search, download, create, and upload research outputs.

Prerequisites

• curl and jq available on PATH.

• For authenticated endpoints (anything under /account/... or uploads), a personal token from https://figshare.com/account/applications exported as:

  export FIGSHARE_TOKEN=xxxxxxxxxxxxxxxx
  

• Public endpoints (search, public articles, downloads) need no token.

Always confirm with the user before creating, modifying, publishing, or deleting anything on their account — these are hard to reverse.

Update check

Throttle to one check per 24 hours per installation; never mutate the skill directory without explicit user consent.

1. If <this-skill-dir>/.last_update exists and is less than 24 hours old, skip this step entirely.

2. Otherwise, fetch the latest tag from upstream:

   git -C <this-skill-dir> ls-remote --tags origin 'v*' 2>/dev/null \
     | awk '{print $2}' | sed 's|refs/tags/||' \
     | sort -V | tail -1
   

3. Compare with this skill's metadata.version from the frontmatter. If the upstream tag is strictly newer (semver), tell the user one line and ask:

   "A newer version of this skill is available: vX.Y.Z → vA.B.C. Want me to git pull?"

   If they say yes, run git -C <this-skill-dir> pull --ff-only. Refresh .last_update either way so the prompt doesn't repeat for 24 hours.

4. If upstream is the same or older, refresh .last_update silently and continue.

5. On any failure (offline, not a git checkout — e.g. ClawHub-installed copy, read-only path, no permission), swallow the error silently and continue with the user's task. Do not mention the failure.

API Basics

• Base URL: https://api.figshare.com/v2
• Auth header: Authorization: token $FIGSHARE_TOKEN
• Content-Type: application/json for POST/PUT bodies
• Rate limit: keep it under ~1 request/second to avoid abuse throttling
• Errors: JSON body with message, code; common codes 400/401/403/404/422

Common Recipes

Search public articles

curl -s -X POST https://api.figshare.com/v2/articles/search \
  -H "Content-Type: application/json" \
  -d '{"search_for": ":title: single cell", "page_size": 20}' | jq

Field operators: :title:, :author:, :tag:, :category:, :doi:, :resource_doi:.

Get a public article (by ID or DOI)

curl -s https://api.figshare.com/v2/articles/{article_id} | jq
# or resolve from a figshare.com URL: the numeric tail is the article_id

Download all files from a public article

ART=12345678
curl -s https://api.figshare.com/v2/articles/$ART/files \
  | jq -r '.[] | "\(.download_url)\t\(.name)"' \
  | while IFS=$'\t' read -r url name; do curl -L -o "$name" "$url"; done

List your own articles

curl -s -H "Authorization: token $FIGSHARE_TOKEN" \
  "https://api.figshare.com/v2/account/articles?page=1&page_size=50" | jq

Create an article (draft)

curl -s -X POST https://api.figshare.com/v2/account/articles \
  -H "Authorization: token $FIGSHARE_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{
    "title": "My dataset",
    "description": "Full description here.",
    "defined_type": "dataset",
    "tags": ["demo"],
    "categories": [2]
  }' | jq

Response is { "location": ".../account/articles/{id}", "entity_id": 123 }.

Update / publish an article

# update metadata
curl -s -X PUT https://api.figshare.com/v2/account/articles/$ART \
  -H "Authorization: token $FIGSHARE_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{"title": "New title"}'

# publish (becomes public, assigns DOI, version is frozen)
curl -s -X POST https://api.figshare.com/v2/account/articles/$ART/publish \
  -H "Authorization: token $FIGSHARE_TOKEN"

Always ask before publishing — it's permanent for that version.

Collections & projects

# create collection that groups existing articles
curl -s -X POST https://api.figshare.com/v2/account/collections \
  -H "Authorization: token $FIGSHARE_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{"title": "My Collection", "articles": [123, 456]}'

# create project
curl -s -X POST https://api.figshare.com/v2/account/projects \
  -H "Authorization: token $FIGSHARE_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{"title": "Research Project"}'

Uploading Files (Multi-part Flow)

Figshare uploads are 3-step: initiate → PUT each part → complete. Use the bundled helpers for anything non-trivial:

# upload a file to an existing draft article
./scripts/upload.sh <article_id> <path/to/file>

# batch-download every file from a public article (accepts id or figshare.com URL)
./scripts/download.sh <article_id_or_url> [output_dir]

# reserve + upload + publish a new version of an already-published article
./scripts/new-version.sh <article_id> <path/to/file>

The raw flow, in case you need to adapt it:

1. Initiate — compute md5 + size, POST to article:

   SIZE=$(stat -f%z "$FILE" 2>/dev/null || stat -c%s "$FILE")
   MD5=$(md5sum "$FILE" | awk '{print $1}')   # or: md5 -q on macOS
   curl -s -X POST https://api.figshare.com/v2/account/articles/$ART/files \
     -H "Authorization: token $FIGSHARE_TOKEN" \
     -H "Content-Type: application/json" \
     -d "{\"md5\":\"$MD5\",\"name\":\"$(basename $FILE)\",\"size\":$SIZE}"
   

   Response has location pointing at /account/articles/$ART/files/$FILE_ID.

2. Fetch upload info from that file record — it contains an upload_url. GET the upload_url to learn the part layout (parts: [{partNo, startOffset, endOffset}]).

3. Upload parts — for each part, PUT the byte range to ${upload_url}/${partNo}:

   dd if="$FILE" bs=1 skip=$START count=$((END-START+1)) 2>/dev/null \
     | curl -s -X PUT --data-binary @- "${upload_url}/${partNo}" \
       -H "Authorization: token $FIGSHARE_TOKEN"
   

4. Complete — POST to the file record to finalize:

   curl -s -X POST https://api.figshare.com/v2/account/articles/$ART/files/$FILE_ID \
     -H "Authorization: token "
   

Why three steps: Figshare streams large files through a separate upload service. Skipping the complete call leaves the file in a pending state and it won't appear on the article.

Pagination

Most list endpoints accept either page+page_size or limit+offset. Max page_size is typically 1000. For large harvests, loop until an empty page:

page=1
while :; do
  out=$(curl -s "https://api.figshare.com/v2/articles?page=$page&page_size=100")
  [ "$(echo "$out" | jq 'length')" = "0" ] && break
  echo "$out" | jq -c '.[]'
  page=$((page+1))
  sleep 1
done

Troubleshooting

• 401 — token missing/expired; re-check $FIGSHARE_TOKEN.
• 403 on /account/... — token lacks the needed scope; regenerate with full permissions.
• 422 on article create — missing required field (usually title) or bad categories/defined_type.
• Upload parts mismatch — md5 or size in step 1 didn't match the bytes actually uploaded; recompute and restart.
• Published article won't update — publishing freezes a version; create a new version instead.

References

• API reference: https://docs.figshare.com/
• Token management: https://figshare.com/account/applications
• Category IDs: GET https://api.figshare.com/v2/categories
• License IDs: GET https://api.figshare.com/v2/licenses