{"id":34520189,"url":"https://github.com/cluesmith/codev","last_synced_at":"2026-05-24T01:05:18.323Z","repository":{"id":315766300,"uuid":"1060250915","full_name":"cluesmith/codev","owner":"cluesmith","description":"Codev helps humans and agents co-develop both the context and the code of the project. ","archived":false,"fork":false,"pushed_at":"2026-05-22T09:46:53.000Z","size":18859,"stargazers_count":271,"open_issues_count":69,"forks_count":36,"subscribers_count":9,"default_branch":"main","last_synced_at":"2026-05-22T10:54:33.328Z","etag":null,"topics":["agentic-ai","agentic-coding","ai-software","software-development"],"latest_commit_sha":null,"homepage":"","language":"TypeScript","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"apache-2.0","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/cluesmith.png","metadata":{"files":{"readme":"README.md","changelog":"CHANGELOG.md","contributing":null,"funding":null,"license":"LICENSE","code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null,"zenodo":null,"notice":null,"maintainers":null,"copyright":null,"agents":"AGENTS.md","dco":null,"cla":null}},"created_at":"2025-09-19T15:54:56.000Z","updated_at":"2026-05-22T01:15:59.000Z","dependencies_parsed_at":"2025-12-24T05:08:26.501Z","dependency_job_id":null,"html_url":"https://github.com/cluesmith/codev","commit_stats":null,"previous_names":["ansari-project/codev"],"tags_count":121,"template":false,"template_full_name":null,"purl":"pkg:github/cluesmith/codev","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/cluesmith%2Fcodev","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/cluesmith%2Fcodev/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/cluesmith%2Fcodev/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/cluesmith%2Fcodev/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/cluesmith","download_url":"https://codeload.github.com/cluesmith/codev/tar.gz/refs/heads/main","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/cluesmith%2Fcodev/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":286080680,"owners_count":33416314,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2026-05-23T22:14:44.296Z","status":"ssl_error","status_checked_at":"2026-05-23T22:14:43.778Z","response_time":53,"last_error":"SSL_read: unexpected eof while reading","robots_txt_status":"success","robots_txt_updated_at":"2025-07-24T06:49:26.215Z","robots_txt_url":"https://github.com/robots.txt","online":false,"can_crawl_api":true,"host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":["agentic-ai","agentic-coding","ai-software","software-development"],"created_at":"2025-12-24T04:46:27.233Z","updated_at":"2026-05-24T01:05:18.291Z","avatar_url":"https://github.com/cluesmith.png","language":"TypeScript","funding_links":[],"categories":[],"sub_categories":[],"readme":"# Codev: A Human-Agent Software Development OS\n\n[![npm version](https://img.shields.io/npm/v/@cluesmith/codev.svg)](https://www.npmjs.com/package/@cluesmith/codev)\n[![License: MIT](https://img.shields.io/badge/License-MIT-yellow.svg)](https://opensource.org/licenses/MIT)\n[![Discord](https://img.shields.io/badge/Discord-Join%20us-5865F2?logo=discord\u0026logoColor=white)](https://discord.gg/mJ92DhDa6n)\n[![LinkedIn](https://img.shields.io/badge/LinkedIn-Follow-0A66C2?logo=linkedin\u0026logoColor=white)](https://www.linkedin.com/company/codev-os/)\n[![Website](https://img.shields.io/badge/Website-codevos.ai-blue)](https://codevos.ai)\n[![Newsletter](https://img.shields.io/badge/Newsletter-Subscribe-green)](https://marketmaker.cluesmith.com/subscribe/codev)\n\n![Agent Farm Dashboard](docs/assets/agent-farm-hero.png)\n\n\u003e **New: [A Tour of CodevOS](https://waleedk.medium.com/a-tour-of-codevos-1db0fe0e4516)** — A deep dive into how Codev orchestrates human-agent collaboration: the SPIR protocol, Agent Farm, multi-model consultation, and the architecture that ties it all together.\n\nCodev turns GitHub Issues into tested, reviewed PRs. You write specs; autonomous AI builders handle the rest.\n\n\u003e **Results**: One architect + autonomous AI builders shipped [106 PRs in 14 days](codev/resources/development-analysis-2026-02-17.md), median feature in 57 minutes. In controlled comparison, SPIR consistently outperformed unstructured AI coding across [4 rounds](codev/resources/vibe-vs-spir-r4-comparison-2026-02.md). [Case study](#-example-implementations) | [Production data](#production-metrics-feb-2026)\n\n**Quick Links**: [FAQ](docs/faq.md) | [Tips](docs/tips.md) | [Cheatsheet](codev/resources/cheatsheet.md) | [CLI Reference](codev/resources/commands/overview.md) | [Why Codev?](docs/why.md) | [Discord](https://discord.gg/mJ92DhDa6n)\n\n📬 **Stay updated** — [Subscribe to the Codev newsletter](https://marketmaker.cluesmith.com/subscribe/codev) for release notes, tips, and community updates.\n\n## Table of Contents\n\n- [Quick Start](#quick-start)\n- [How It Works](#how-it-works)\n- [Learn About Codev](#learn-about-codev)\n- [What is Codev?](#what-is-codev)\n- [The SPIR Protocol](#the-spir-protocol)\n- [Example Implementations](#-example-implementations)\n- [Real-World Performance](#-eating-our-own-dog-food)\n- [Agent Farm](#agent-farm-optional)\n- [Contributing](#contributing)\n\n## Quick Start\n\n```bash\n# 1. Install\nnpm install -g @cluesmith/codev\n\n# 2. Initialize a project\nmkdir my-project \u0026\u0026 cd my-project\ncodev init\n\n# 3. Verify setup\ncodev doctor\n\n# 4. Start the workspace (optional)\nafx workspace start\n```\n\nThen open a GitHub Issue describing what you want to build, and run:\n\n```bash\nafx spawn \u003cissue-number\u003e\n```\n\nFor the full walkthrough, see **[Getting Started](https://codevos.ai/getting-started)**.\n\n**CLI Commands:**\n- `codev` - Main CLI (init, adopt, doctor, update)\n- `afx` - Agent Farm for parallel AI builders\n- `consult` - Multi-model consultation\n\nSee [CLI Reference](codev/resources/commands/overview.md) for details.\n\n## How It Works\n\n1. **Write a spec** — Describe what you want. The architect helps refine it.\n2. **Spawn a builder** — `afx spawn 42` kicks off an autonomous agent in an isolated worktree.\n3. **Review the plan** — The builder writes an implementation plan. You approve or annotate.\n4. **Walk away** — The builder implements, tests, and opens a PR. You review and merge.\n\n### Prerequisites\n\n**Core (required):**\n\n| Dependency | Install | Purpose |\n|------------|---------|---------|\n| Node.js 18+ | `brew install node` | Runtime |\n| git 2.5+ | (pre-installed) | Version control |\n| AI CLIs | See below | All three recommended |\n\n**AI CLIs** (install all three for multi-model consultation):\n- Claude Code: `npm install -g @anthropic-ai/claude-code`\n- Gemini CLI: [github.com/google-gemini/gemini-cli](https://github.com/google-gemini/gemini-cli)\n- Codex CLI: `npm install -g @openai/codex`\n\n**Agent Farm (optional):**\n\n| Dependency | Install | Purpose |\n|------------|---------|---------|\n| gh | `brew install gh` | GitHub CLI |\n\nSee [DEPENDENCIES.md](codev-skeleton/DEPENDENCIES.md) for complete details. \n\n## Learn about Codev\n\n### ❓ FAQ\n\nCommon questions about Codev: **[FAQ](docs/faq.md)**\n\n### 💡 Tips \u0026 Tricks\n\nPractical tips for getting the most out of Codev: **[Tips \u0026 Tricks](docs/tips.md)**\n\n### 📋 Cheatsheet\n\nQuick reference for Codev's philosophies, concepts, and tools: **[Cheatsheet](codev/resources/cheatsheet.md)**\n\n### 📺 Quick Introduction (5 minutes)\n[![Codev Introduction](https://img.youtube.com/vi/vq_dmfyMHRA/0.jpg)](https://youtu.be/vq_dmfyMHRA)\n\nWatch a brief overview of what Codev is and how it works.\n\n*Generated using [NotebookLM](https://notebooklm.google.com/notebook/e8055d06-869a-40e0-ab76-81ecbfebd634) - Visit the notebook to ask questions about Codev and learn more.*\n\n### 💬 Participate\n\nJoin the conversation in [GitHub Discussions](https://github.com/cluesmith/codev/discussions) or our [Discord community](https://discord.gg/mJ92DhDa6n)! Share your specs, ask questions, and learn from the community.\n\n**Get notified of new discussions**: Click the **Watch** button at the top of this repo → **Custom** → check **Discussions**.\n\n### 📺 Extended Overview (Full Version)\n[![Codev Extended Overview](https://img.youtube.com/vi/8KTHoh4Q6ww/0.jpg)](https://www.youtube.com/watch?v=8KTHoh4Q6ww)\n\nA comprehensive walkthrough of the Codev methodology and its benefits.\n\n### 🛠️ Agent Farm Demo: Building a Feature with AI\n[![Agent Farm Demo](https://img.youtube.com/vi/0OEhdk7-plE/0.jpg)](https://www.youtube.com/watch?v=0OEhdk7-plE)\n\nWatch a real development session using Agent Farm - from spec to merged PR in 30 minutes. Demonstrates the Architect-Builder pattern with multi-model consultation.\n\n### 🎯 Codev Tour - Building a Conversational Todo Manager\nSee Codev in action! Follow along as we use the SPIR protocol to build a conversational todo list manager from scratch:\n\n👉 [**Codev Demo Tour**](https://github.com/ansari-project/codev-demo/blob/main/codev-tour.md)\n\nThis tour demonstrates:\n- How to write specifications that capture all requirements\n- How the planning phase breaks work into manageable chunks\n- The implementation phase in action\n- Multi-agent consultation with GPT-5 and Gemini Pro\n- How lessons learned improve future development\n\n## What is Codev?\n\nCodev is a development methodology that treats **natural language context as code**. Instead of writing code first and documenting later, you start with clear specifications that both humans and AI agents can understand and execute.\n\n📖 **Read the full story**: [Why We Created Codev: From Theory to Practice](docs/why.md) - Learn about our journey from theory to implementation and how we built a todo app without directly editing code.\n\n### Core Philosophy\n\n1. **Context Drives Code** - Context definitions flow from high-level specifications down to implementation details\n2. **Human-AI Collaboration** - Designed for seamless cooperation between developers and AI agents\n3. **Evolving Methodology** - The process itself evolves and improves with each project\n\n## The SPIR Protocol\n\nOur flagship protocol for structured development:\n\n- **S**pecify - Define what to build in clear, unambiguous language\n- **P**lan - Break specifications into executable phases\n- **I**mplement - Build the code, write tests, verify requirements for each phase\n- **R**eview - Capture lessons and improve the methodology\n\n## Project Structure\n\nAfter running `codev init` or `codev adopt`, your project has a **minimal structure**:\n\n```\nyour-project/\n├── codev/\n│   ├── specs/              # Feature specifications\n│   ├── plans/              # Implementation plans\n│   ├── reviews/            # Review and lessons learned\n│   └── resources/           # Reference materials\n├── AGENTS.md               # AI agent instructions (AGENTS.md standard)\n├── CLAUDE.md               # AI agent instructions (Claude Code)\n└── [your code]\n```\n\n### Customizable and Extendable\n\nCodev is designed to be customized for your project's needs. The `codev/` directory is yours to extend:\n\n- **Add project-specific protocols** - For example, Codev itself has a `release` protocol specific to npm publishing\n- **Customize existing protocols** - Modify SPIR phases to match your team's workflow\n- **Add new roles** - Define specialized consultant or reviewer roles\n\nThe framework provides defaults, but your local files always take precedence.\n\n### Context Hierarchy\n\nIn much the same way an operating system has a memory hierarchy, Codev repos have a context hierarchy. The codev/ directory holds the top 3 layers. This allows both humans and agents to think about problems at different levels of detail.\n\n![Context Hierarchy](codev/resources/context-hierarchy.png)\n\n**Key insight**: We build from the top down, and we propagate information from the bottom up. We start with a GitHub issue, then spec and plan out the feature, generate the code, and then propagate what we learned through the reviews.\n\n## Key Features\n\n### 📄 Natural Language is the Primary Programming Language\n- Specifications and plans drive implementation\n- All decisions captured in version control\n- Clear traceability from idea to implementation\n\n### 🤖 AI-Native Workflow\n- Structured formats that AI agents understand\n- Multi-agent consultation support (GPT-5, Gemini Pro, etc.)\n- Reduces back-and-forth from dozens of messages to 3-4 document reviews\n- Supports both AGENTS.md standard (Cursor, Copilot, etc.) and CLAUDE.md (Claude Code)\n\n### 🔄 Continuous Improvement\n- Every project improves the methodology\n- Lessons learned feed back into the process\n- Templates evolve based on real experience\n\n## 📚 Example Implementations\n\nBoth projects below were given **the exact same prompt** to build a Todo Manager application using **Claude Code with Opus**. The difference? The methodology used:\n\n### [Todo Manager - VIBE](https://github.com/ansari-project/todo-manager-vibe)\n- Built using a **VIBE-style prompt** approach (same model, same prompt)\n- Produced boilerplate scaffolding but 0% of the specified functionality\n- No tests, no database, no working API — demonstrates how conversational approaches can miss the mark entirely\n\n### [Todo Manager - SPIR](https://github.com/ansari-project/codev-demo)\n- Built using the **SPIR protocol** with full document-driven development\n- Same requirements, but structured through formal specifications and plans\n- Demonstrates all phases: Specify → Plan → Implement → Review\n- Complete with specs, plans, and review documents\n- Multi-agent consultation throughout the process\n\n\u003cdetails\u003e\n\u003csummary\u003e\u003cstrong\u003e📊 Multi-Agent Comparison (4 rounds)\u003c/strong\u003e (click to expand)\u003c/summary\u003e\n\n**Methodology**: Same prompt, same AI model (Claude Opus). Unstructured (conversational) vs SPIR (structured protocol). Scored by 3 independent AI agents (Claude, Codex, Gemini Pro) on a 1-10 scale. Full auto-approved gates — no human review input — to isolate the protocol's effect.\n\n#### Latest Results (Round 4, Feb 2026)\n\n| Dimension | Unstructured | SPIR | Delta |\n|-----------|:----------:|:----:|:-----:|\n| **Overall** | **5.8** | **7.0** | **+1.2** |\n| Bugs | 6.7 | 7.3 | +0.7 |\n| Code Quality | 7.0 | 7.7 | +0.7 |\n| Tests | 5.0 | 6.7 | +1.7 |\n| Deployment | 2.7 | 6.7 | +4.0 |\n\n#### Key Findings\n\n- **+1.2 quality advantage consistent across all 4 rounds** (R1: +1.3, R2: +1.2, R4: +1.2)\n- SPIR produced **2.9x more test code** with broader layer coverage\n- SPIR produced **fewer source lines** (1,249 vs 1,294) while being more complete — the first round where structured code was more concise\n- **Deployment readiness** showed the largest delta of any dimension in any round (+4.0): multi-stage Dockerfile, standalone output, deploy instructions\n- Multi-agent consultation caught **5 implementation bugs pre-merge** at a cost of $4.38\n\n**Build time**: SPIR took ~56 min vs ~15 min for unstructured (3.7x). Consultation accounts for 45% of the overhead. Estimated cost: $14-19 vs $4-7 (3-5x). For production code, the deployment readiness and test coverage alone justify the investment.\n\nSee [full Round 4 analysis](codev/resources/vibe-vs-spir-r4-comparison-2026-02.md) for detailed scoring, bug sweeps, and architecture comparison.\n\n\u003c/details\u003e\n\n## 🐕 Eating Our Own Dog Food\n\nCodev is **self-hosted** — we use Codev to build Codev. Every feature goes through SPIR. Every improvement has a spec, plan, and review.\n\n### Production Metrics (Feb 2026)\n\nOver a 14-day sprint building Codev with Codev ([full analysis](codev/resources/development-analysis-2026-02-17.md)):\n\n| Metric | Value |\n|--------|-------|\n| Merged PRs | 106 |\n| Closed issues | 105 |\n| Commits | 801 |\n| Median feature implementation | 57 minutes |\n| Fully autonomous builders | 85% (22 of 26) |\n| Pre-merge bugs caught by consultation | 20 |\n| Consultation cost per PR | $1.59 |\n\nOne architect with autonomous builders matched the output of a **3-4 person elite engineering team** (benchmarked against 5 PRs/developer/week from LinearB's 2026 analysis of 8.1M PRs). The bugfix pipeline is genuinely autonomous: 66% of fixes ship in under 30 minutes (median 13 min from PR creation to merge).\n\nMulti-agent consultation catches real bugs that single-model review misses. No single reviewer found all 20 bugs — Codex excels at edge-case exhaustiveness, Claude at runtime semantics, Gemini at architecture.\n\nThis self-hosting approach ensures:\n1. The methodology is battle-tested on real development\n2. We experience the same workflow we recommend to users\n3. Pain points are felt by us first and fixed quickly\n4. The framework evolves based on actual usage, not theory\n\n### Understanding This Repository's Structure\n\nThis repository has a dual nature:\n\n1. **`codev/`** - Our instance of Codev for developing Codev itself\n   - Contains our specs, plans, reviews, and resources\n   - Example: `codev/specs/0001-test-infrastructure.md` documents how we built our test suite\n\n2. **`codev-skeleton/`** - The template that gets installed in other projects\n   - Contains protocol definitions, templates, and agents\n   - What users get when they install Codev\n   - Does NOT contain specs/plans/reviews (those are created by users)\n\n**In short**: `codev/` is how we use Codev, `codev-skeleton/` is what we provide to others.\n\n\u003cdetails\u003e\n\u003csummary\u003e\u003cstrong\u003eTest Infrastructure\u003c/strong\u003e (click to expand)\u003c/summary\u003e\n\nOur test suite validates the Codev CLI and Agent Farm:\n\n- **Framework**: Vitest (unit tests) + Playwright (E2E tests)\n- **Coverage**: CLI commands, porch protocol orchestration, Agent Farm dashboard\n- **Isolation**: Tests run in isolated environments\n\n```bash\n# From packages/codev/\nnpm test              # Run unit tests\nnpm run test:e2e      # Run E2E tests\n```\n\nSee [Testing Guide](codev/resources/testing-guide.md) for details.\n\n\u003c/details\u003e\n\n## Examples\n\n### Todo Manager Tutorial\n\nSee `examples/todo-manager/` for a complete walkthrough showing:\n- How specifications capture all requirements\n- How plans break work into phases\n- How the implementation phase ensures quality\n- How lessons improve future development\n\n## Configuration\n\n### Customizing Templates\n\nTemplates in `codev/protocols/spir/templates/` can be modified to fit your team's needs:\n\n- `spec.md` - Specification structure\n- `plan.md` - Planning format\n- `lessons.md` - Retrospective template\n\n## Agent Farm (Optional)\n\nAgent Farm is an optional companion tool for Codev that provides a web-based dashboard for managing multiple AI agents working in parallel. **You can use Codev without Agent Farm** - all protocols (SPIR, TICK, etc.) work perfectly in any AI coding assistant.\n\n**Why use Agent Farm?**\n- **Web dashboard** for monitoring multiple builders at once\n- **Protocol-aware** - knows about specs, plans, and Codev conventions\n- **Git worktree management** - isolates each builder's changes\n- **Automatic prompting** - builders start with instructions to implement their assigned spec\n\n**Current limitations:**\n- Uses **shellper processes** for persistent terminal sessions (node-pty handles terminal I/O)\n- macOS-focused (should work on Linux but less tested)\n\n### Alternative Agent Shells\n\nAgent Farm supports multiple AI coding agents as shells. Configure in `.codev/config.json`:\n\n**Claude Code** (default):\n```json\n{\n  \"shell\": {\n    \"architect\": \"claude --dangerously-skip-permissions\",\n    \"builder\": \"claude --dangerously-skip-permissions\"\n  }\n}\n```\n\n**OpenCode** (builder only):\n```json\n{\n  \"shell\": {\n    \"architect\": \"claude --dangerously-skip-permissions\",\n    \"builder\": \"opencode run\"\n  }\n}\n```\n\nOpenCode supports 75+ LLM providers (OpenAI, Anthropic, Google, Ollama, etc.). When using OpenCode as a builder:\n- Include `run` in the command (plain `opencode` launches the TUI, which hangs in a PTY session)\n- Configure tool permissions for unattended execution in `~/.config/opencode/opencode.json` or your project's `opencode.json`:\n  ```json\n  { \"permission\": { \"edit\": \"allow\", \"bash\": \"allow\" } }\n  ```\n- OpenCode reads `AGENTS.md` for project instructions (already present in Codev projects)\n- OpenCode is only supported as a **builder** shell, not as an architect shell\n\nOther shells (Codex, Gemini) are also supported via the harness system. See `packages/codev/src/agent-farm/utils/harness.ts` for details, or define a custom harness in `.codev/config.json`.\n\n## Architect-Builder Pattern\n\nFor parallel AI-assisted development, Codev includes the Architect-Builder pattern:\n\n- **Architect** (you + primary AI): Creates specs and plans, reviews work\n- **Builders** (autonomous AI agents): Implement specs in isolated git worktrees\n\n### Quick Start\n\n```bash\n# Start the workspace\nafx workspace start\n\n# Spawn a builder for a spec\nafx spawn 3 --protocol spir\n\n# Check status\nafx status\n\n# Stop everything\nafx workspace stop\n```\n\nThe `afx` command is globally available after installing `@cluesmith/codev`.\n\n### Remote Access\n\nAccess Agent Farm from any device via cloud connectivity:\n\n```bash\nafx tower connect\n```\n\nRegister your tower with [codevos.ai](https://codevos.ai) for secure remote access from any browser — no SSH tunnels or port forwarding needed.\n\n### Autonomous Builder Flags\n\nBuilders need permission-skipping flags to run autonomously without human approval prompts:\n\n| CLI Tool | Flag | Purpose |\n|----------|------|---------|\n| Claude Code | `--dangerously-skip-permissions` | Skip permission prompts for file/command operations |\n| Gemini CLI | `--yolo` | Enable autonomous mode without confirmations |\n\nConfigure in `.codev/config.json` (created by `codev init` or `codev adopt`):\n```json\n{\n  \"shell\": {\n    \"architect\": \"claude --dangerously-skip-permissions\",\n    \"builder\": \"claude --dangerously-skip-permissions\"\n  }\n}\n```\n\nOr for Gemini:\n```json\n{\n  \"shell\": {\n    \"architect\": \"gemini --yolo\",\n    \"builder\": \"gemini --yolo\"\n  }\n}\n```\n\n**Warning**: These flags allow the AI to execute commands and modify files without asking. Only use in development environments where you trust the AI's actions.\n\nSee [CLI Reference](codev/resources/commands/agent-farm.md) for full documentation.\n\n## Releases\n\nCodev has a **release protocol** (`codev/protocols/release/`) that automates the entire release process. To release a new version:\n\n```\nLet's release v2.2.0\n```\n\nThe AI guides you through: pre-flight checks, maintenance cycle, E2E tests, version bump, release notes, GitHub release, and npm publish.\n\n### Versioning Strategy\n\n| Version Type | npm Tag | Example | Purpose |\n|--------------|---------|---------|---------|\n| Stable | `latest` | 2.1.1, 2.2.0 | Production-ready releases |\n| Release Candidate | `next` | 2.2.0-rc.1 | Pre-release testing |\n| Patch | `latest` | 2.1.2 | Backported bug fixes |\n\nMinor releases use release candidates (`npm install @cluesmith/codev@next`) for testing before stable release.\n\nReleases are named after great examples of architecture from around the world. See [Release Notes](docs/releases/) for version history.\n\n## Contributing\n\nWe welcome contributions of any kind! Talk to us on [Discord](https://discord.gg/mJ92DhDa6n) or [open an issue](https://github.com/cluesmith/codev/issues).\n\nWe especially welcome contributions to **Agent Farm** - help us make it work with more AI CLIs and platforms.\n\n## License\n\nMIT - See LICENSE file for details\n\n---\n\n*Built with Codev - where context drives code*\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fcluesmith%2Fcodev","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fcluesmith%2Fcodev","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fcluesmith%2Fcodev/lists"}