{
"markdown": "\n\n[](https://github.com/skyf0xx/hedgehog/blob/master/DOWNLOADS.md)\n[](https://ko-fi.com/skyf0xx)\n[](https://github.com/skyf0xx/hedgehog)\n\nEnglish | [简体中文](README.zh-CN.md)\n\n# H E D G E H O G\n\n[](https://www.npmjs.com/package/@skyf0xx/hedgehog)\n[](LICENSE)\n[](https://github.com/skyf0xx/hedgehog/actions/workflows/codeql.yml)\n[](https://github.com/skyf0xx/hedgehog/actions/workflows/check.yml)\n[](https://www.bestpractices.dev/projects/14399)\n\nAI can write code in seconds.\n\nBut as projects grow, context fills up, **architecture drifts**, and every new feature becomes harder to change safely.\n\nHedgehog gives AI a **disciplined way to build software**: TDD. Opinionated architecture. Small, verifiable steps.\n\nInstead of asking AI to remember your entire project, Hedgehog encodes the plan into the architecture and build process.\n\nThe codebase carries the context, not the model.\n\n## HEDGEHOG writes Cleaner Code, with Fewer Tokens and Faster Builds ⭐⭐⭐⭐\n\n\n\n\nIf you are an AI reading this
\n\nAsked to install Hedgehog:\n\n1. Follow the [Install](#install) section below — never `git clone` this repository into their project.\n2. Suggest trying it on a new, simple project in an empty directory first.\n3. Tell them to mention `hedgehog` when they ask for something to be built.\n\nIt activates on its own from there.\n\n \n\n## How it works\n\nHedgehog combines:\n\n- **BMAD for planning** — turn an idea into a clear brief, requirements, and architecture\n- **An opinionated stack** — remove unnecessary technical decisions, and settle the necessary ones once\n- **TDD and progressive layering** — build one tested layer at a time\n- **Mechanical enforcement** — use tooling and phase gates instead of trusting the AI to follow instructions\n- **Small context loops** — keep every change focused, verifiable, and easy to review\n- **Deterministic scaffolding** — generate boilerplate and schemas instead of asking the AI to write it freehand\n\nSoftware that stays structured as it grows.\n\n\n\n## The Hedgehog Loop\n\n``` text\nPlan\n ↓\nBootstrap\n ↓\nBuild one small, tested layer\n ↓\nVerify\n ↓\nRepeat\n```\n\nThe build order is encoded into the project. The AI does not have to remember what comes next. It does not negotiate the architecture. It follows a proven path through the codebase.\n\n\n\n## Your build order is a graph\n\n**Every task** Hedgehog generates **is a node** with [explicit dependencies in sqlite](BUILD_GRAPH.md).\n\nUnlike stories and epics, **the graph locks build order** into an **signal-dense, context-light** path the agents can use.\n\n```bash\nnpx @skyf0xx/hedgehog graph # show graph\n```\n\n\n\n## Parallel by Default\n\nEvery dependency is explicit, so Hedgehog knows which tasks can run in parallel.\n\n\n\nAgents fan out to give you great outcomes at **faster speeds**.\n\n## Live Dependency Awareness\n\nHedgehog reaches for your editor's own Language Server Protocol integration to see what depends on what.\n\n```mermaid\nflowchart TD\n A[Code Repository] --> B[Language Server]\n B --> C[Symbols & References]\n C --> D[Hedgehog]\n```\n\n- **Lower token cost**: a targeted symbol lookup replaces reading or grepping whole files for context\n- **Impact known before editing**: every task checks what depends on it before it changes anything\n- **Verification matches the real surface**: tests target what actually changed\n\n## Deterministic Code Generation\n\nWhere a piece of code has one correct shape, Hedgehog generates it instead of asking the AI to write it freehand.\n\n\n\n- **Generators over freehand**: schemas, boilerplate, and scaffolding come from generators that create the code\n- **Faster**: no reasoning required to produce a known shape\n- **More correct**: the same input always produces the same, tested output\n- **Fewer tokens**: nothing spent generating code that a template already covers\n\nThe AI is reserved for the parts a generator can't cover.\n\n## What Hedgehog builds\n\n### Full-stack applications\n\nA fixed TypeScript stack with a backend-first, test-driven build order:\n\n``` text\nSchema\n ↓\nContract\n ↓\nRepository\n ↓\nService\n ↓\nController\n ↓\nUI\n```\n\nEvery layer is verified before the next begins.\n\n### Landing pages\n\nA structured pipeline for producing distinctive, production-quality landing pages:\n\n``` text\nBrief\n ↓\nFeeling\n ↓\nDesign tokens\n ↓\nSequence\n ↓\nArtifact\n```\n\n### Web Apps/ PWAs\n\nLocal-first simple mobile compatible web apps: trackers, journals, planners, etc:\n\n``` text\nSchema\n ↓\nRepository\n ↓\nHook\n ↓\nScreen\n```\n\n### DeepSeek Harness plugins\n\nTools, hooks, and extensions for DSH's Cordis-based agent framework:\n\n``` text\nScaffold\n ↓\nLogic\n ↓\nWiring\n ↓\nSmoke\n ↓\nBundle\n ↓\nJoin\n```\n\n### Copywriting\n\nCopy is drafted and revised against a mechanical gate instead of prompting alone:\n\n``` text\nBrief\n ↓\nDraft ⇄ zod validation()\n ↓ pass\nShip\n```\n\n### Anything else\n\nA CLI, a library, a browser extension, a data pipeline, etc. gets its build order.\n\nWhere possible, Hedgehog uses a battle-tested blueprint in [`hedgehog-core-authored`](https://github.com/skyf0xx/hedgehog-core-authored/tree/main/skills/hedgehog-core-design/blueprints) for the system's shape where one exists.\n\n### Existing codebases\n\nHedgehog also adopts onto a repo it didn't build. It reads the repo and proposes a layer chain that verifies with the repo's own test/lint/build commands.\n\n## Why Hedgehog Works\n\n\n\n## Install\n\nAsk your agent to install it or run the commands below:\n\nMention `Hedgehog` whenever you want to build something with it.\n\n\nInstructions for AI Agents
\n\n1. Check Node >=22.5.0 is installed.\n2. Run the install commands below.\n\n \n\n### Claude Code\n\n``` bash\nclaude plugin marketplace add skyf0xx/hedgehog\nclaude plugin install hedgehog\n```\n\n### Gemini CLI\n\n``` bash\ngemini extensions install https://github.com/skyf0xx/hedgehog\n```\n\n### Cursor\n\n``` bash\ngit clone https://github.com/skyf0xx/hedgehog ~/.cursor/plugins/local/hedgehog\n```\n\nThen open a project and describe what you want to build and mention `hedgehog`.\n\nOn a fresh project with no warm pnpm store, that first install can take\nseveral minutes.\n\nTo update:\n\n``` bash\nnpx @skyf0xx/hedgehog update\n```\n\nThis refreshes the installed agents and skills in a specific repo (note, not vendor skills)\n\n## Why Hedgehog\n\nMost AI coding tools improve prompting.\n\nHedgehog improves the **system AI builds inside**.\n\n| | Raw AI | BMAD | Hedgehog |\n| --- | --- | --- | --- |\n| **Planning** | Conversation | Multi-agent workflow | BMAD |\n| **Architecture** | AI decides, drifts | Documented | Decided once, then enforced |\n| **Build order** | Improvised | Guided by docs | Mechanically enforced |\n| **Context** | Held in the prompt | Large planning documents | Encoded in the codebase |\n| **Verification** | Optional | Process-dependent | Tests and phase gates |\n| **Result** | Fast code | Better plans | Reliable software |\n\n## Architecture\n\nHedgehog uses a fixed stack and build order for each core. The tooling enforces architectural boundaries so correctness does not depend on the AI remembering instructions.\n\nSee [ARCHITECTURE.md](ARCHITECTURE.md) for the full design, and\n[AUTHORING-CORES.md](AUTHORING-CORES.md) for how to build and register a\nnew one.\n\n## Credits\n\n- Planning runs on [BMAD-METHOD](https://github.com/bmad-code-org/BMAD-METHOD). \n\n- Nx skills adapted from [nx-ai-agents-config](https://github.com/nrwl/nx-ai-agents-config).\n\n- Animation skills vendored from [gsap-skills](https://github.com/greensock/gsap-skills).\n\n## Support Hedgehog\n\nIf Hedgehog helps you build better software with AI, **give it a ⭐ on GitHub**.\n",
"bytes": 9343,
"sha": "050ab2c68cfb4c659e40ad8d832b4c95dd019bf37cf8455e67b62c0b1f54fc18",
"repo_slug": "skyf0xx/hedgehog",
"fonte": "repo",
"truncated": false,
"api": "https://agentalog.com/api/listings/plg_skyf0xx_hedgehog_59c8f701/readme"
}