Context-Engineering
starsforks
How to install Context-Engineering?
apoo711/context-engineering
$git clone https://github.com/Apoo711/Context-EngineeringA spec-driven development toolkit to eliminate AI code bloat and automate software lifecycle tasks.
README.md
| 1 | # 🚀 Context Engineering Suite (Antigravity CLI Edition) |
| 2 | |
| 3 | A spec-driven development toolkit to eliminate AI code bloat and automate software lifecycle tasks. |
| 4 | |
| 5 | --- |
| 6 | |
| 7 | 💡 **Foundational Philosophy:** Read about our philosophy in the blog post: [**Beyond the Prompt: A Practical Guide to Context Engineering with Gemini**](https://aryan-gupta.is-a.dev/blog/2026-06-18-cxt-eng). |
| 8 | |
| 9 | --- |
| 10 | |
| 11 | ## ⚡ Why Context Engineering 2.0? |
| 12 | |
| 13 | Context Engineering 2.0 represents a complete paradigm shift from raw, conversational prompts to structured software engineering workflows. Instead of "vibe-coding" and hoping the LLM gets it right, Context Engineering structures the context, templates, rules, and commands natively within the **Antigravity CLI (agy)** to ensure consistent, reliable, and production-grade code generation. |
| 14 | |
| 15 | ### Comparison: Vibe Coding vs. Context Engineering 2.0 |
| 16 | |
| 17 | | Feature | Vibe Coding (Prompt Engineering 1.0) | Context Engineering 2.0 (Antigravity Edition) | |
| 18 | | :--- | :--- | :--- | |
| 19 | | **Workflow** | Sending raw, unstructured prompts in chat. | Executing compiled, structured specifications (PRPs). | |
| 20 | | **Code Quality** | Prone to over-engineered blocks and dependency bloat. | Minimalist, spec-driven code restricted by toggleable guidelines (Ponytail mode). | |
| 21 | | **Interaction** | Constant conversational fluff and back-and-forth. | Native, non-interactive action blocks and precise file diffs. | |
| 22 | | **Context Management** | Context drift as the conversation grows. | Unified, project-specific rulesets and slash commands natively registered in `agy`. | |
| 23 | |
| 24 | --- |
| 25 | |
| 26 | ## 📦 Quick Installation |
| 27 | |
| 28 | Install the unified plugin natively in your terminal with a single command: |
| 29 | |
| 30 | ```shell |
| 31 | agy plugin install https://github.com/Apoo711/Context-Engineering |
| 32 | ``` |
| 33 | |
| 34 | --- |
| 35 | |
| 36 | ## 🛠️ The Core Toolkit (Commands) |
| 37 | |
| 38 | The suite registers two native slash commands directly into your `agy` environment to manage the lifecycle of your features: |
| 39 | |
| 40 | ### 1. `/generate-prp` |
| 41 | Generates a rigid, comprehensive **Product Requirements Prompt (PRP)** template based on a simple user request. It forces the AI to plan edge cases, testing gates, and structures before writing a single line of code. |
| 42 | |
| 43 | ### 2. `/execute-prp` |
| 44 | Executes a compiled PRP spec sheet. It uses strict, non-interactive action blocks to cleanly modify or create files without conversational fluff. |
| 45 | |
| 46 | --- |
| 47 | |
| 48 | ## 💇 Toggleable Ponytail Mode (Minimalist Coding) |
| 49 | |
| 50 | Tired of AI agents pulling down heavy third-party libraries for trivial tasks? **Ponytail Mode** is our built-in minimalist coding guidelines ruleset. |
| 51 | |
| 52 | The philosophy is simple: **"Lazy, not negligent."** Ponytail mode forces the agent to exhaustively evaluate standard libraries and native browser features before introducing external dependencies. |
| 53 | |
| 54 | ### How to Enable |
| 55 | You can configure Ponytail Mode by editing the `enable_ponytail_mode` setting in your `plugin.json`: |
| 56 | |
| 57 | ```json |
| 58 | { |
| 59 | "settings": { |
| 60 | "enable_ponytail_mode": true |
| 61 | } |
| 62 | } |
| 63 | ``` |
| 64 | |
| 65 | Alternatively, you can toggle it interactively through the visual `/config` menu in `agy` or update it in your global settings. |
| 66 | |
| 67 | --- |
| 68 | |
| 69 | ## 🗺️ Roadmap |
| 70 | |
| 71 | Here is where the Context Engineering Suite is heading next: |
| 72 | |
| 73 | - **Progressive Agent Personas:** Toggle between specialized agent profiles (e.g., Greenfield Builder vs. Hardened-Security Auditor). |
| 74 | - **Auto-Verification Test Hooks:** Post-invocation validation gates that automatically run your test suites to verify generated code correctness. |
| 75 | - **Visual Workspace Indexing (`/context-map`):** Auto-generate interactive dependency and architecture maps directly in your terminal. |
| 76 | |
| 77 | --- |
| 78 | |
| 79 | ## 🤝 Acknowledgements |
| 80 | |
| 81 | This project's workflow and foundational concepts were inspired by the original [Context-Engineering-Intro](https://github.com/coleam00/context-engineering-intro) repository by **coleam00**. While the codebase has transitioned into a native Antigravity CLI plugin, the core vision of spec-driven AI development remains. |
People who liked this also viewed
6 picksType
- components-helperBased on the docs to provide code prompt files for vue component libraryPromptsNov 2023
- VibeCode-Protocol-SuiteA comprehensive suite of protocols, meta-prompts, and orchestration tools designed to streamline software development workflows, project management, and team collaboration. Includes the VibeCode Protocol Suite with web interface, user manuals, and reusable templates for efficient development processes.PromptsJul 2026
- Promptable-web-sdkWeb SDK for Promptable Website.PromptsDec 2022
- LLMFlexA python package for developing AI applications with local LLMs.PromptsJan 2025
- cli-repo-to-promptexport any codebase to a single LLM-ready markdown promptPromptsJul 2026
- Files2PromptVS Code extension to copy files for an LLM promptPromptsApr 2025
