biobankjapan-phewas-skill — Codex skill: features and setup
Retrieves single-variant BioBank Japan PheWAS summaries from rsID, GRCh37, or GRCh38 inputs. Useful for rapid genetic association checks inside your coding agent.
npx skills add openai/plugins --skill biobankjapan-phewas-skill -a codexWhat biobankjapan-phewas-skill does
The skill operates by calling scripts/biobankjapan_phewas.py through standard input. It expects a JSON object containing exactly one variant key: rsid, grch37, grch38, or variant. Before contacting BioBank Japan, it normalizes coordinates into the required GRCh37 chr:pos-ref-alt string.
Responses render as compact Markdown summaries by default. The tool recommends starting with max_results=10 to avoid consuming unnecessary context tokens. Whenever the complete dataset is needed, setting save_raw=true writes raw records to disk and reports the file location via raw_output_path.
Execution requires a local Python runtime alongside the scripts/biobankjapan_phewas.py helper script.
What it helps with
- Retrieve BioBank Japan PheWAS summary statistics for a given rsID
- Map GRCh38 genomic coordinates to canonical GRCh37 before querying BBJ
- Export large association arrays to disk without overflowing context
- Inspect variant-phenotype links directly within an agent terminal
Example prompts for the agent
- Look up BioBank Japan PheWAS associations for rs7903146 with the first 10 hits
- Query BBJ for GRCh38 coordinate
10:112998590-C-Tand save raw output to disk - Run a PheWAS search for variant
10:114758349:C:Twith max 25 results
How to install biobankjapan-phewas-skill
1Skills CLI (Vercel)
npx skills add openai/plugins --skill biobankjapan-phewas-skill -a codex2Manually: put the files in place
.agents/skills/biobankjapan-phewas-skill/SKILL.mdFor all projects use ~/.agents/skills/
1Skills CLI (Vercel)
npx skills add openai/plugins --skill biobankjapan-phewas-skill -a claude-code2Manually: put the files in place
~/.claude/skills/biobankjapan-phewas-skill/SKILL.mdCopy the whole skill folder; for one project use .claude/skills/biobankjapan-phewas-skill/
Manually: put the files in place
.cursor/skills/biobankjapan-phewas-skill/SKILL.mdCursor also reads .agents/skills/ and .claude/skills/
1GitHub CLI
gh skill install openai/plugins biobankjapan-phewas-skill2Manually: put the files in place
.github/skills/biobankjapan-phewas-skill/SKILL.mdPlatform CLI
gemini skills install https://github.com/openai/plugins --path plugins/life-science-research/skills/biobankjapan-phewas-skillManually: put the files in place
.devin/skills/biobankjapan-phewas-skill/SKILL.mdThe legacy .windsurf/skills/ path still works
Claude.ai
Zip the folder and upload it in Settings → Capabilities → Skills
Step by step, with commands for each platform: Codex skills: how to install, where they live, how to call them
Questions about biobankjapan-phewas-skill
How do I install biobankjapan-phewas-skill in Codex?
Run npx skills add openai/plugins --skill biobankjapan-phewas-skill. This registers the skill within your agent environment.
What input parameters does biobankjapan-phewas-skill accept?
It requires exactly one variant identifier: rsid, grch37, grch38, or variant. Optional keys include max_results, save_raw, raw_output_path, and timeout_sec.
How does the skill prevent large outputs from cluttering chat?
It displays markdown summaries limited by max_results. Setting save_raw=true writes the full payload to disk and outputs raw_output_path.
