| name | skill-template |
| description | Load when copying this directory to bootstrap a new OmicsClaw v2 skill (rename, fill in, then `git add`). Skip when an existing skill already covers the request. |
| version | 0.1.0 |
| author | OmicsClaw |
| license | MIT |
| tags | ["template","scaffold","v2"] |
REPLACE_SKILL_NAME
When to use
The user has <input shape> and wants <output shape>. Pick this skill
when <distinguishing condition>. For <adjacent capability> use
<sibling-skill> instead.
Inputs & Outputs
| Input | Format | Required |
|---|
| Primary input | <.h5ad / .csv / .vcf / …> | yes (unless --demo) |
| Output | Path | Notes |
|---|
| Primary table | tables/<name>.csv | one row per <unit> |
| Report | report.md + result.json | always |
Flow
- Load input (
--input <file>) or generate a demo (--demo).
- Validate required columns /
obs[X] keys; raise ValueError(...) early.
- Run the chosen
--method backend.
- Write
tables/<name>.csv (<script>.py:<L>) + report.md + result.json.
Gotchas
- None yet — append as failure modes are reported.
Key CLI
python omicsclaw.py run REPLACE_SKILL_NAME --demo --output /tmp/REPLACE_SKILL_NAME_demo
python omicsclaw.py run REPLACE_SKILL_NAME \
--input <data.ext> --output results/ \
--method <method-name>
See also
references/parameters.md — every CLI flag, per-method tunablesreferences/methodology.md — the WHY behind the algorithmreferences/output_contract.md — tables/X.csv + result.json schema- Adjacent skills:
<sibling-1> (upstream), <sibling-2> (parallel), <sibling-3> (downstream)
