Skip to content

context-engineering-kit

by NeoLabHQ

Hand-crafted Claude Code Skills focused on improving agent results quality. Compatible with OpenCode, Cursor, Antigravity, Gemini CLI, and others.

About

Hand-crafted Claude Code Skills focused on improving agent results quality. Compatible with OpenCode, Cursor, Antigravity, Gemini CLI, and others.

728
Stars
86
Skill Files
62
Forks
55
PRs Merged
5
Contributors
5
Issues Open
11
Issues Closed
2026-03-29
Last Activity

Skill Analysis

563
Avg Lines / Skill
3,169
Avg Words / Skill
1146
Code Blocks
299
Total Files
3.37 MB
Repo Size
0
Reference Files
Skill lines16 min / 391 median / 6068 max
Code languagesbash, dot, gitconfig, javascript, json, jsonc, language, markdown, mermaid, python, suggestion, text, typescript, xml, yaml
Frontmatter keysallowed-tools, argument-hint, color, description, model, name, tools
Has CLAUDE.mdYes
Has LicenseYes
File typesmd (255), json (19), sh (7), ts (6), yaml (1), lockb (1), png (1)

Skills (86)

  • git:commit
    187 lines1381 words1 blocks

    Create well-formatted commits with conventional commit messages and emoji

  • git:worktrees
    527 lines2099 words31 blocks

    Use when working on multiple branches simultaneously, context switching without stashing, reviewing PRs while developing, testing in isolation, or com...

    bash
  • git:create-pr
    144 lines647 words7 blocks

    Create pull requests using GitHub CLI with proper templates and formatting

    bash
  • git:merge-worktree
    320 lines1606 words16 blocks

    Merge changes from worktrees into current branch with selective file checkout, cherry-picking, interactive patch selection, or manual merge

    bash
  • git:attach-review-to-pr
    512 lines1876 words25 blocks

    Add line-specific review comments to pull requests using GitHub CLI API

    bashjsonmarkdown
  • git:create-worktree
    257 lines1199 words15 blocks

    Create and setup git worktrees for parallel development with automatic dependency installation

    bash
  • git:analyze-issue
    81 lines331 words

    Analyze a GitHub issue and create a detailed technical specification

  • git:compare-worktrees
    286 lines1198 words20 blocks

    Compare files and directories between git worktrees or worktree and current branch

    bash
  • git:load-issues
    55 lines219 words1 blocks

    Load all open issues from GitHub and save them as markdown files

    markdown
  • git:notes
    438 lines1480 words24 blocks

    Use when adding metadata to commits without changing history, tracking review status, test results, code quality annotations, or supplementing commit ...

    bashgitconfig
  • mcp:setup-arxiv-mcp
    117 lines583 words5 blocks

    Guide for setup arXiv paper search MCP server using Docker MCP

    markdowntext
  • mcp:build-mcp
    359 lines1787 words1 blocks

    Guide for creating high-quality MCP (Model Context Protocol) servers that enable LLMs to interact with external services through well-designed tools. ...

    xml
  • mcp:setup-codemap-cli
    244 lines886 words11 blocks

    Guide for setup Codemap CLI for intelligent codebase visualization and navigation

    bashjsonmarkdowntext
  • mcp:setup-context7-mcp
    57 lines270 words2 blocks

    Guide for setup Context7 MCP server to load documentation for specific technologies.

    markdowntext
  • mcp:setup-serena-mcp
    121 lines689 words4 blocks

    Guide for setup Serena MCP server for semantic code retrieval and editing capabilities

    markdowntext
  • Setup TypeScript best practices and code style rules in CLAUDE.md

    markdown
  • reflexion:memorize
    304 lines1474 words4 blocks

    Curates insights from reflections and critiques into CLAUDE.md using Agentic Context Engineering

    bashmarkdownyaml
  • reflexion:reflect
    651 lines3243 words7 blocks

    Reflect on previus response and output, based on Self-refinement framework for iterative improvement with complexity triage and verification

    javascriptmarkdown
  • reflexion:critique
    478 lines1734 words8 blocks

    Comprehensive multi-perspective review using specialized judges with debate and consensus building

    bashmarkdown
  • meta-judge
    762 lines5142 words13 blocks

    Use this agent when generating evaluation rubrics, checklists, criteria, metrics, and weights for a user prompt BEFORE implementation begins. Produces...

    markdownyaml

Recent Commits

-346 contributions across all repos
MarAprMayJunJulAugSepOctNovDecJanFebMar
Less
More
  • 48d6d482026-03-29
    Merge pull request #72 from NeoLabHQ/fix/vg/code-review-and-do-in-parallel-skills
    Vladislav Goncharov
  • 74f43212026-03-29
    fix(sadd): make do in parallel perform meta-judge evaluation per task
    leovs09
  • 5f2a7dc2026-03-29
    fix(code-reivew): remove local code review tools limitations
    leovs09
  • d2bb7a82026-03-22
    Merge pull request #70 from NeoLabHQ/chore/remove-outdated-docker-scripts
    Vladislav Goncharov
  • 4f7db102026-03-22
    chore: remove docker outside of docker script
    leovs09
  • 6d2dec22026-03-21
    feat(sadd): add LLM-as-a-Meta-Judge rubrics generation step
    Vladislav Goncharov
  • 0453f1f2026-03-21
    docs(code-review): add arguments decription for commands
    leovs09
  • 889aa4e2026-03-21
    docs: add deeper description of how rules work
    leovs09
  • 0fd1b5c2026-03-21
    feat(sadd): use 5 why for rule creation analysis diring judging
    leovs09
  • 347d1ea2026-03-21
    docs(ddd): update docs with new rules description
    leovs09