Skip to content

docs(pixi-skill): refresh pixi-package-manager skill to pixi 0.72.1#117

Merged
lsetiawan merged 1 commit into
mainfrom
docs/pixi-skill-refresh-0.72.1
Jul 8, 2026
Merged

docs(pixi-skill): refresh pixi-package-manager skill to pixi 0.72.1#117
lsetiawan merged 1 commit into
mainfrom
docs/pixi-skill-refresh-0.72.1

Conversation

@lsetiawan

Copy link
Copy Markdown
Member

Summary

  • Refresh the pixi-package-manager skill from pixi 0.69.0 → 0.72.1.
  • Every documented command was verified against the installed 0.72.1 binary and cross-checked against the official docs (llms-full.txt).
  • Fixes two commands that no longer exist in 0.72.1, adds newly-available features, and removes duplication per a skill-authoring review.

Changes

Correctness (commands that fail in 0.72.1):

  • pixi add numpy -c conda-forge --force-reinstallpixi add "conda-forge::numpy" (both -c and --force-reinstall error in 0.72.1)
  • pixi tree --depth 2pixi tree / pixi tree -i numpy (--depth removed; -i/--invert is current)
  • Windows install iwr -useb … | iexirm -useb … | iex to match the official docs

New capabilities (since 0.69.0):

  • requires-pixi minimum-version pin added to the reproducibility checklist and the reproducible-research pattern
  • Lockfile archival modernized to the first-class pixi workspace export conda-environment / conda-explicit-spec commands

Metadata & authoring cleanup:

  • Frontmatter: pixi-version: 0.72.1, last-verified: 2026-07-08
  • Collapsed the duplicated global/exec/shell-hook concept into a pointer to the Quick Reference Card
  • Kept the SHA-pinning rationale in a single source (best-practices reference + asset header); SKILL.md now points instead of restating
  • Removed the empty scripts/ directory

Test plan

  • pixi --version reports 0.72.1 (baseline for verification)
  • Every command in SKILL.md and references/ parses/runs on pixi 0.72.1
  • pixi add "conda-forge::numpy", pixi tree -i <pkg>, pixi workspace requires-pixi set, and pixi workspace export conda-environment all succeed
  • All TOML snippets parse (Pattern 5 requires-pixi under [tool.pixi.workspace])

Bump 0.69.0 -> 0.72.1, with every command verified against the installed
0.72.1 binary and cross-checked against the official docs (llms-full.txt).

- Frontmatter: pixi-version 0.72.1, last-verified 2026-07-08
- Replace two commands that no longer exist in 0.72.1:
  - `pixi add -c conda-forge --force-reinstall` -> `pixi add "conda-forge::numpy"`
  - `pixi tree --depth 2` -> `pixi tree` / `pixi tree -i` (--depth removed)
- Windows install: `iwr` -> `irm` to match the official docs
- Add `requires-pixi` version pin (new since 0.69.0) to the reproducibility
  checklist and the reproducible-research pattern
- Modernize lockfile archival to `pixi workspace export conda-environment` /
  `conda-explicit-spec` (first-class in 0.72.1)
- Tighten per a skill-authoring review: collapse the duplicated
  global/exec/shell-hook concept into a pointer, keep the SHA-pinning
  rationale in a single source, drop the empty scripts/ dir
@github-actions

github-actions Bot commented Jul 8, 2026

Copy link
Copy Markdown

🔍 Tessl Skill Review

plugins/scientific-python-development/skills/pixi-package-manager/SKILL.md

⚠️ Error: ✘ Skill review requires you to be logged in. Run tessl login to log in, then try again.

Note: tessl skill review is deprecated – use tessl review run instead.
Migration guide: https://docs\.tessl\.io/improving\-your\-skills/migrate\-from\-skill\-review


To improve your score, point your agent at the Tessl optimization guide. Need help? Jump on our Discord.

Feedback

Report issues with this review at tesslio/skill-review, or send private feedback from your terminal with tessl feedback.

@lsetiawan
lsetiawan merged commit 1b59a87 into main Jul 8, 2026
2 of 3 checks passed
@lsetiawan
lsetiawan deleted the docs/pixi-skill-refresh-0.72.1 branch July 8, 2026 17:36
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant