agentleFS
SYipSYip/dirban/.agents/skills/code-review/SKILL.md

code-review

Skill0 starsChanged 2 days agoMITView on GitHub

Perform systematic, rigorous code reviews for diffs, pull requests, staged changes, or recent commits. Verifies correctness, architecture compliance, algorithmic complexity (Big O analysis & O(n²) flags), design patterns for maintainability, TypeScript standards (FP, immutability, no semicolons, single concern, named constants/regexes), test vector coverage, and security invariants.

194 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use close-spec, design-with-images and 10 more.

Other files people use alongside this one.

Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
  • dzhngSkill

    close-spec

    When a spec under specs/<feature>/ is done shipping (a write-spec build, or any planned task that produced a spec), archive it to specs/done/ and rewrite it from a build-plan into a durable rationale — the why, the principles, the invariants — pointing back to the real code for the how. Use when implementation has landed and the plan no longer matches what shipped, or the user says a feature/spec is finished. Pairs with [write-spec](../write-spec/SKILL.md) (the plan this closes) and [review](../review/SKILL.md) (run before closing).

    Also in 5 projects that use this

  • dzhngSkill

    design-with-images

    Explore UI and visual design with image generation, then iterate the real implementation against the user-selected concept. Use when brainstorming visual directions, offering mockup variants, or implementing an approved generated design; do not stop at the mockups or the first code approximation.

    Also in 5 projects that use this

  • PydanticSkill

    add-new-model

    Add support for a newly-released language or image generation model in pydantic-ai (e.g. openai:gpt-5.6, anthropic:claude-sonnet-5, openai:gpt-image-2). Use when a provider ships a new model id and you need to wire literals, profile flags, adapters, and tests to recognize it. Handles SDK-lag, gateway list conventions, capability probing, and direct image-model geometry.

    Also in 9 projects that use this

  • transitive-bullshitSkill

    compare-screenshots

    Compare screenshots against the intended design, distinguishing approved references from historical baselines. Use for iterative implementation matching, before/after visual review, objective image telemetry, or checking a lone capture for flat, empty, or badly framed content.

    Also in 5 projects that use this

  • Microsoft AzureSkill

    azsdk-common-apiview-feedback-resolution

    Analyze and resolve APIView review feedback on Azure SDK PRs. **UTILITY SKILL**. USE FOR: APIView comments, API review feedback, SDK API surface changes. DO NOT USE FOR: general code review, non-APIView feedback. INVOKES: azure-sdk-mcp:azsdk_apiview_get_comments, azure-sdk-mcp:azsdk_customized_code_update, azure-sdk-mcp:azsdk_typespec_delegate_apiview_feedback, azure-sdk-mcp:azsdk_run_typespec_validation, azure-sdk-mcp:azsdk_package_generate_code.

    Also in 6 projects that use this

  • Microsoft AzureSkill

    azsdk-common-generate-sdk-locally

    Generate and finalize Azure SDKs locally from TypeSpec, including build, checks, tests, changelog, metadata, and version updates by default. WHEN: "generate SDK locally", "build SDK", "run SDK tests", "run CI checks", "validate package", "run checks", "update changelog", "fix SDK build errors", "resolve SDK generation errors", "customize TypeSpec", "rename SDK client", "rename SDK model", "hide operation from SDK", "fix analyzer errors", "resolve customization drift", "create subclient", "update metadata", "update version". INVOKES: azsdk_verify_setup, azsdk_package_generate_code, azsdk_package_build_code, azsdk_package_run_check, azsdk_package_run_tests, azsdk_customized_code_update, azsdk_package_update_changelog_content, azsdk_package_update_metadata, azsdk_package_update_version.

    Also in 6 projects that use this

  • +4 more from dzhng
  • +2 more from Azure

The file

We split this skill into the rules it gives an agent

12 rules. 0 are common, 9 are distinctive, and 3 are very specific to this project.

  • 0

    Common

    Most of the 184 edited versions we read keep these.

  • 9

    Distinctive

    This author's own choices. Worth a look.

  • 3

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Run code review before completing tasks, creating pull requests, or when asked to inspect changes.

    line 18
  2. Distinctive:

    If reviewing work on an active task, inspect its acceptance criteria and run lint, typecheck, and unit tests:

    line 38
  3. Distinctive:

    Check if a function or utility already exists before writing new ones.

    line 58
  4. Distinctive:

    Strategy Pattern: When branching logic or runtime variants are hardcoded in sprawling conditionals (e.g.

    line 64
  5. Very specific:

    Adapter / Facade Pattern: When wrapping raw external third-party APIs or Node.js primitives into pure, neverthrow-safe domain interfaces.

    line 65
  6. Distinctive:

    Factory / Builder Pattern: When constructing complex data structures or domain models with numerous optional parameters or invariant …

    line 66
  7. Distinctive:

    Pipeline / Chain of Responsibility: When processing data through multi-step transformations, validations, or sanitizations.

    line 68
  8. Distinctive:

    If quadratic behavior is strictly unavoidable or operating on a tiny, fixed-size constant array (e.g.

    line 81
  9. Very specific:

    Do your best to prevent nested levels of if blocks.

    line 86
  10. Distinctive:

    Check if any complex logic, calculations, or validations can be extracted into cohesive, pure helper functions.

    line 89
  11. Very specific:

    No anonymous inline regex literals (e.g.

    line 93
  12. Distinctive:

    No uncaught throwing in core domain logic.

    line 104

“Common” is measured: kept by at least 6 in 10 of the 184 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

Code Review Skill

This skill guides agents and developers through a structured, objective code review of uncommitted changes, branches, or PR diffs against repository standards.


1. Review Triggers & Input Inspection

Run code review before completing tasks, creating pull requests, or when asked to inspect changes.

Step 1: Gather the Diff

Determine what changes are being reviewed:

# Unstaged working tree changes
git diff

# Staged changes ready for commit
git diff --cached

# Changes against base branch (main)
git diff main...HEAD

# Recent commit
git diff HEAD~1..HEAD

Step 2: Check Active Task & Context

If reviewing work on an active task, inspect its acceptance criteria and run lint, typecheck, and unit tests:

dirban current --raw
bun run check
bun run typecheck
bun test

2. Review Checklist & Evaluation Dimensions

Evaluate the diff against these 7 core dimensions:

A. Architectural & Scope Principles

  • Review Scope & Focus:
    • Focus strictly on modified code and touched functions in the diff.
    • Ignore untouched existing code; do not flag or expand scope to clean up legacy code unless the function itself was modified.
  • Check Existing Logic & Pragmatic Reuse:
    • Check if a function or utility already exists before writing new ones.
    • Reusing existing logic is preferred, but intentional duplication is acceptable when it avoids premature abstractions, avoids unnecessary coupling across distinct boundaries, or keeps a module self-contained.
  • Progressive Enhancement: Does the code keep filesystem-as-state primary? Is Git / external tooling strictly optional?
  • No Ball-of-Mud: Are files, modules, and functions cohesive? Does each component have a single responsibility?
  • Design Pattern Recommendations:
    • Suggest proven design patterns that elevate code quality, decouplability, and testability:
      • Strategy Pattern: When branching logic or runtime variants are hardcoded in sprawling conditionals (e.g. interchangeable VCS providers, workspace lifecycles, formatting engines).
      • Adapter / Facade Pattern: When wrapping raw external third-party APIs or Node.js primitives into pure, neverthrow-safe domain interfaces.
      • Factory / Builder Pattern: When constructing complex data structures or domain models with numerous optional parameters or invariant validation.
      • Command / Dispatch Table: Replace large switch/case branches with typed lookup dictionaries or command handler maps.
      • Pipeline / Chain of Responsibility: When processing data through multi-step transformations, validations, or sanitizations.
    • Caution: Do not recommend patterns that introduce unnecessary abstraction layers or over-engineering for simple, one-off logic.
  • Documentation & Rule Integrity: Are documentation and docstrings preserved?

B. Algorithmic Complexity & Big-O Analysis

  • Big-O Complexity Check:
    • Evaluate time and space complexity of new or modified loops, lookups, and data transformations.
    • Flag any $O(n^2)$ or worse quadratic patterns:
      • Nested iterations over collections (e.g. items.map(i => otherItems.find(o => o.id === i.id))).
      • Calling Array.prototype.includes, find, indexOf, or filter inside loops over large or unbounded arrays.
      • Repeated array allocations, concatenations ([...acc, item]), or string slicing inside tight loops.
    • Remediation:
      • Pre-index collections into Map or Set ($O(1)$ lookup) before the loop to reduce complexity to $O(n)$.
      • If quadratic behavior is strictly unavoidable or operating on a tiny, fixed-size constant array (e.g. 3 columns), explicitly document the bound and rationale.

C. TypeScript Standards (.agents/rules/typescript.md)

  • No Semicolons: Code uses semi: false (no trailing semicolons).
  • Flat Control Flow & Early Returns:
    • Do your best to prevent nested levels of if blocks.
    • Invert conditions and use early return, continue, or break (guard clauses) to keep control flow flat wherever practical.
  • Logic Extraction:
    • Check if any complex logic, calculations, or validations can be extracted into cohesive, pure helper functions.
  • Named Magic Values & Complex Booleans:
    • Magic strings (column names, commands, status values) and numbers (timeouts, widths, limits) are defined as named constants (const / as const).
    • Complex boolean conditions (with multiple operators or non-trivial logic) are extracted into descriptive const is... = ... variables or predicate functions.
    • No anonymous inline regex literals (e.g. const EXTRACT_REGEX = /.../).
  • Functional & Immutable:
    • Prefers compile-time readonly types over noisy runtime Object.freeze wrapping.
    • Uses map, filter, reduce, Object.fromEntries instead of mutating arrays/objects.
    • Interfaces, parameters, and array return types use readonly T[] / Readonly<T>.
    • No push, splice, or in-place property mutations.
  • Single Responsibility & Function Scope:
    • Functions do one thing with one level of abstraction.
    • Long switchboards or sprawling multi-concern functions are refactored into smaller helpers or dispatch tables.
  • Error Handling with neverthrow:
    • Domain operations return Result<T, E> (ok / err).
    • No uncaught throwing in core domain logic.
  • Strict Typing:
    • Strict types with no any or untyped casts.
    • Null checks prefer == null / != null to catch both null and undefined.
  • No Single-Letter Variable Names:
    • Descriptive, self-documenting variable names (e.g., use query, parsedItem, taskIndex, NOT q, p, x). The only permitted single-letter variable is i in numeric index loops.

D. Testing & Verification (.agents/rules/tests.md)

  • Test Vectors: Are new behaviors covered by explicit test vectors (Vector 1, Vector 2, ...)?
  • Clean Suite: Do all tests pass?
    bun test
  • Typecheck: Does the compiler report zero errors?
    bun run typecheck

E. File & Code Organization (.agents/rules/file-naming.md)

  • Kebab-Case: Directory and file names are kebab-case (e.g., task-mutation.ts, types.ts).
  • PascalCase: Classes, React/TUI components, and types use PascalCase.
  • Test Suffix: Tests use .test.ts placed adjacent or in matching directory.

F. Safety, Security & Git Constraints

  • No Unprompted Commits: Never run git commit or git reset unless explicitly commanded by user.
  • Clean Resource Handling: File descriptors, spawned child processes, and temporary directories are safely cleaned up.

3. Output Format

Present code review findings in a structured, actionable format:

### Summary
A concise 1-2 sentence overview of the changeset and overall health.

### 🔴 Critical Issues (Blockers)
Issues that cause bugs, break invariants, or fail tests. Include file link and line numbers.

### 🟡 Complexity & Algorithmic Improvements (Big-O & Performance)
Inefficient logic, redundant passes, unnecessary allocations or copying, unindexed lookups (e.g., pre-indexing into `Set`/`Map`), or O(n²) / quadratic patterns.

### 🟡 Design Patterns & Architectural Improvements
Concrete design pattern suggestions (e.g. Strategy, Adapter, Dispatch Map, Pipeline) that simplify conditionals, decouple dependencies, or improve testability.

### 🟡 Improvements & Standards (Non-blocking)
TypeScript rule violations, immutability gaps, named regex suggestions, or refactoring opportunities.

### 🟢 Strengths & Highlights
Well-architected patterns, good test coverage, or clean type safety.

### Verification Status
- Tests: `bun test` (X passed, 0 failed)
- Typecheck: `bun run typecheck` (clean / errors)

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.