cellxgene-skill — Codex skill: metadata and install
This skill sends compact CELLxGENE Discover API requests and summarizes public collection and dataset metadata. Use it for focused single-cell lookups.
npx skills add openai/plugins --skill cellxgene-skill -a codexWhat cellxgene-skill does
cellxgene-skill calls the public CELLxGENE Discover API through scripts/rest_request.py. It reads one JSON object from stdin with base_url and path, then returns compact records or a summary. Use collections/<collection_id> for a focused lookup; use collections when the task calls for broad archive discovery.
The skill can pass query parameters, headers, request bodies, item limits, and depth limits. Raw responses can be saved with save_raw=true and raw_output_path. The default result is concise Markdown, while raw JSON is returned only when requested. A large collections listing may need a higher timeout_sec, so a targeted request is the safer starting point for that case.
The package contains SKILL.md and scripts/rest_request.py. Its error response includes ok=false, an error code, and a message, which gives the agent a clear check before parsing the result.
What it helps with
- look up metadata for a CELLxGENE collection
- summarize a public single-cell dataset
- search the public CELLxGENE collection list
- save a raw API response for later inspection
Example prompts for the agent
- Use `cellxgene-skill` to fetch concise metadata for CELLxGENE collection `db468083-041c-41ca-8f6f-bf991a070adf`.
- Use `cellxgene-skill` to search public CELLxGENE collections and return the first `5` records.
- Query CELLxGENE through `scripts/rest_request.py`, save the raw response, and return a compact summary.
How to install cellxgene-skill
1Skills CLI (Vercel)
npx skills add openai/plugins --skill cellxgene-skill -a codex2Manually: put the files in place
.agents/skills/cellxgene-skill/SKILL.mdFor all projects use ~/.agents/skills/
1Skills CLI (Vercel)
npx skills add openai/plugins --skill cellxgene-skill -a claude-code2Manually: put the files in place
~/.claude/skills/cellxgene-skill/SKILL.mdCopy the whole skill folder; for one project use .claude/skills/cellxgene-skill/
Manually: put the files in place
.cursor/skills/cellxgene-skill/SKILL.mdCursor also reads .agents/skills/ and .claude/skills/
1GitHub CLI
gh skill install openai/plugins cellxgene-skill2Manually: put the files in place
.github/skills/cellxgene-skill/SKILL.mdPlatform CLI
gemini skills install https://github.com/openai/plugins --path plugins/life-science-research/skills/cellxgene-skillManually: put the files in place
.devin/skills/cellxgene-skill/SKILL.mdThe legacy .windsurf/skills/ path still works
Claude.ai
Zip the folder and upload it in Settings → Capabilities → Skills
From the same set201
The whole set →Similar
Questions about cellxgene-skill
How do I install `cellxgene-skill` in Codex?
Run npx skills add openai/plugins --skill cellxgene-skill, then run npx skills add openai/plugins --skill cellxgene-skill -a codex. The installed file is available at .agents/skills/cellxgene-skill/SKILL.md.
What does `cellxgene-skill` do?
It sends requests to the public CELLxGENE Discover API and produces concise collection and dataset metadata. It favors focused collections/<collection_id> requests and returns compact Markdown by default.
What should I know before requesting the CELLxGENE collection list?
The collections path can produce a large result, so it may need a higher timeout_sec. A focused collection request is usually the better first call.
