agentleFS
NumericalEarthNumericalEarth/Breeze.jl/.claude/skills/add-feature/SKILL.md

add-feature

Skill90 starsChanged 3 months agoApache-2.0View on GitHub

Checklist for adding new physics or features to Breeze

27 other projects use a version of this, 1 of them edited.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

9 rules. 0 are common, 2 are distinctive, and 7 are very specific to this project.

  • 0

    Common

    Most of the 25 edited versions we read keep these.

  • 2

    Distinctive

    This author's own choices. Worth a look.

  • 7

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Follow this checklist when adding new physics or features to Breeze.

    line 8
  2. Very specific:

    All structs concretely typed.

    line 14
  3. Very specific:

    Use materialization pattern: skeleton struct → materialize_* for fully-typed version

    line 15
  4. Very specific:

    Use @kernel and @index from KernelAbstractions.jl

    line 17
  5. Distinctive:

    Keep type-stable and allocation-free

    line 18
  6. Very specific:

    Use ifelse instead of short-circuiting if/else

    line 19
  7. Very specific:

    No loops over grid points outside kernels — use launch!

    line 21
  8. Very specific:

    Verify on CPU: Run tests with ENV["CUDA_VISIBLE_DEVICES"] = "-1"

    line 25
  9. Very specific:

    Check explicit imports: Run quality_assurance test

    line 26

“Common” is measured: kept by at least 6 in 10 of the 25 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Add Feature

Follow this checklist when adding new physics or features to Breeze.

Checklist

  1. Create module in the appropriate subdirectory under src/
  2. Define types/structs with docstrings (use jldoctest blocks, never plain julia)
    • All structs concretely typed. Never use Any as a type parameter.
    • Use materialization pattern: skeleton struct → materialize_* for fully-typed version
  3. Implement kernel functions — must be GPU-compatible:
    • Use @kernel and @index from KernelAbstractions.jl
    • Keep type-stable and allocation-free
    • Use ifelse instead of short-circuiting if/else
    • Mark inner functions with @inline
    • No loops over grid points outside kernels — use launch!
  4. Add unit tests in test/ (ParallelTestRunner autodiscovery will find them)
  5. Update exports in src/Breeze.jl if the user interface changed
  6. Add validation example in examples/ when appropriate
  7. Verify on CPU: Run tests with ENV["CUDA_VISIBLE_DEVICES"] = "-1"
  8. Check explicit imports: Run quality_assurance test

Key Conventions

  • File names: snake_case
  • Type names: PascalCase
  • Function names: snake_case
  • Kernel names: may be prefixed with underscore (e.g., _compute_tendency_kernel)
  • Extend functions via Module.function_name(...), never via import

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.