galaxy-ground-truth-expansion
Purpose: expand each benchmark item’s acceptable ground-truth tools (multiple correct answers) by manually reviewing the query and the underlying tutorial context.'
npx skills add majiayu000/claude-skill-registry --skill galaxy-ground-truth-expansion --agent claude-code
Same command for any agent — swap --agent for codex, cursor, copilot.
Weekly change comes from our own snapshots, not the repository page — it measures attention, not adoption.
# Galaxy Ground-Truth Expansion (Manual, per item) Purpose: expand each benchmark item’s **acceptable ground-truth tools** (multiple correct answers) by **manually** reviewing the query and the underlying tutorial context. This skill exists because automated, “template-like” expansions can easily introduce incorrect alternatives. This skill is for maintaining `data/benchmark/v1_items.jsonl` and keeping `data/benchmark/v1_items_readable.md` in sync. ## Non-negotiable rules - Expansion is **manual**: review **one query + one tool** at a time. - Do **not** use “pattern rules”, “keyword triggers”, or “template expansions” to add tools in bulk. - Only add an alternative tool if it is **clearly acceptable** for the same user intent **in Galaxy**. - Only add tool IDs that exist on the target server (usegalaxy.org) tool universe. - Prefer a small, high-precision set of alternatives over a large noisy set. ## What “acceptable alternative” means An alternative tool is acceptable if, given the query as written: - A typical Galaxy user could reasonably choose either tool to accomplish the task, and - The tool is not a different *analysis step* (e.g., a preprocessing step vs a downstream step),
- Non-negotiable rules
- What “acceptable alternative” means
- Inputs / outputs
- Why IO details + help text matter (discovery + validation)
- Step 0 — Make sure you have the tool universe
- Step 1 — Pick an item to expand
- Step 2 — Propose alternatives (manual)
- Step 3 — Apply the change (edit JSONL)
- Step 4 — Re-export the readable view
- Step 5 — Spot-check
- Notes
- Worked example — SVM classifier → Tabular Learner
- Discovery process
- Decision
What does the galaxy-ground-truth-expansion skill do?
Purpose: expand each benchmark item’s acceptable ground-truth tools (multiple correct answers) by manually reviewing the query and the underlying tutorial context.'
How do I install it?
Run `npx skills add majiayu000/claude-skill-registry --skill galaxy-ground-truth-expansion --agent claude-code` — it drops the skill into your project so the agent can pick it up. Swap the --agent value for codex, cursor or copilot if you use one of those.
Where does this skill come from?
From majiayu000/claude-skill-registry, a repository with 534 stars. We read it straight from the repository tree rather than a submitted listing, so what you see here is what is actually published.
Is a popular skill a good skill?
Not necessarily. Stars measure attention, not adoption — a repository can trend for a week and be abandoned. That is why we show the weekly change from our own snapshots next to the total, instead of a single flattering number.