modules/home/programs/cli-agents/shared/skills/coding/code-style/SKILL.md
Apply Not Matthias's Rust-first personal coding style. Use whenever the user explicitly asks to apply or review their code style, make Rust match their preferences, perform a style pass, or simplify/refactor according to their conventions. Inspect only task-touched code, honor local project conventions first, and make only safe opt-out style edits.
npx skillsauth add not-matthias/dotfiles-nix code-styleInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
Use this skill only for an explicit request to apply, review, or refactor toward the user's code style. Rust is the primary language. For another language, use only the generic fallback below and follow that language's project conventions.
Read references/rust-style.md before judging a Rust change. It is a standalone personal rule ledger: it contains the rules, rationale, examples, confidence, and application boundaries needed for this skill. Do not depend on or mention outside repositories when using it.
Resolve conflicts in this order:
A local convention wins. Do not call it a violation or migrate it toward the reference. Mention it only when the distinction explains why no patch is appropriate.
cognitive-load, minimal-diff, and testing guidance when it applies. This skill selects personal preferences; it does not replace those workflows.Applying or Recommended:
review-only, no edits, show the patch, or equivalent, apply eligible auto-apply changes in the same task.Apply a rule only when the change is local, behavior-preserving, non-public, and does not change drop timing, ownership, error behavior, allocation behavior, or concurrency.
Always propose instead of automatically editing when a change affects:
Result/Option semantics, error variants, retry behavior, or logging policy;Drop, allocation, concurrency, or performance-sensitive behavior;unsafe, FFI/ABI layout, platform behavior, or safety invariants;If the evidence or semantic equivalence is uncertain, leave the code unchanged and explain the smallest safe next step.
Use the reference to favor:
?, let else, and explicit match where state or dispatch matters;These are preferences, not license to introduce systems, FFI, or performance ceremony into ordinary Rust.
Outside Rust, apply only universal, low-risk improvements when local conventions support them:
Do not impose Rust module structure, error types, ownership patterns, logging, or formatting on another language. Treat cross-file renames, API changes, and any behavioral uncertainty as proposals.
Use this structure when it adds value:
Applying
1. [area] smallest change — why
Recommended, not applied
1. [risk · area] change — why it needs confirmation
Kept local convention
1. local pattern — why it overrides the reference
Check
- command or scenario actually run
If no material style change is warranted, say so plainly. Do not manufacture findings to make a style pass look productive.
Update references/rust-style.md only when the user explicitly asks to evolve their style. Keep it self-contained: add the rule, rationale, a generic example when useful, confidence/application boundary, and exceptions. Do not silently learn rules from one task or rely on external project sources.
documentation
Save notes, journal entries, and research to the personal-notes Obsidian vault (personal-vault-v2). Use when the user asks to 'save note', 'save to notes', 'write to personal notes', 'save to daily notes', 'note this down', or wants to persist findings/analysis to their personal vault.
documentation
Use whenever the user asks to address, fix, resolve, review, or respond to pull-request comments or review feedback.
development
Guide for writing ast-grep rules to perform structural code search and analysis. Use when users need to search codebases using Abstract Syntax Tree (AST) patterns, find specific code structures, or perform complex code queries that go beyond simple text search. This skill should be used when users ask to search for code patterns, find specific language constructs, or locate code with particular structural characteristics.
testing
Plan a huge chunk of work — more than one agent session can hold — as a shared map of decision tickets on your issue tracker, and resolve them one at a time until the way to the destination is clear.