Makes your AI agent think like the laziest senior dev in the room. The best code is the code you never wrote.
AI coding assistant for VS Code, JetBrains, and CLI
CoStrict packages an AI coding assistant as editor extensions, a CLI tool, and related services. Its main pieces are strict mode for guided generation, repository-wide code review, code completion, and MCP integration for tools and data.
Builders who want a coding assistant that follows stricter workflows across editor and CLI setups.
You can generate, review, and refine code with more structured steps instead of ad hoc prompts.
What it does
Strict mode
Guides AI code generation through requirements analysis, architecture design, task planning, and test generation.
Code review
Indexes the repository and uses RAG-based analysis with multi-model verification.
Code completion
Generates context-aware code in seconds inside the editor or CLI flow.
MCP integration
Connects the assistant to APIs, databases, and custom tools through standardized integrations.
Multi-modal input
Accepts image context and other visual inputs alongside code and text.
Skills support
Adds extensible skills for specialized task workflows.
Private deployment
Supports enterprise deployment with isolation and encrypted handling for internal use.
README
CoStrict is a free, open-source AI-powered coding assistant designed for enterprise-grade development. With support for private deployment, it's the optimal choice for organizations requiring secure, standardized AI development workflows.
✨ Core Capabilities
| Feature | Description |
|---|---|
| 🔒 Strict Mode | Standardized AI code generation with requirements analysis, architecture design, task planning, and test generation |
| 🔍 Code Review | Repository-wide RAG-based code analysis with multi-expert model verification |
| ⚡ Code Completion | Context-aware code generation in seconds |
| 🎯 Vibe Code | Rapid development through natural language dialogue |
| 🔗 MCP Integration | Standardized system connectivity for APIs, databases, and custom tools |
| 🎨 Multi-modal | Support for image context and visual inputs |
| 🛠️ Skills Support | Extensible skills system for specialized task workflows |
📦 Installation
|
🌟 CoStrict Cloud |
VS Code Extension |
CLI Tool |
JetBrains Plugin |
🚀 Key Features
Strict Mode
Standardizes AI-generated code workflows for enterprise scenarios, ensuring high-quality and controllable outputs.
|
|
Code Review
Repository-wide indexing and RAG-based analysis with multi-model verification strategies.
|
|
More Features
- 🌐 Multi-language Support - Python, Go, Java, JavaScript/TypeScript, C/C++, and all programming languages
- 🔐 Privacy & Security - Professional private deployment with physical isolation and end-to-end encryption
- 🎛️ API & Model Customization - Built-in free models + support for Anthropic, OpenAI, OpenAI-compatible APIs, and local models
- 📁 Large Repository Context - Automatic context inclusion with @ file/folder mentions
- 🔧 Mode Customization - Multiple default modes (Code, Orchestrator) with custom mode support
- 📝 OpenSpec Integration - Standardized change proposal workflows with
/openspec-init - 🔗 VSCode SCM Integration - Seamless integration with Git for code review and collaboration
- Auto Cleanup History - Automatically clean up history and reduce disk usage.
- 📝 History Import&Export - Supports import and export of historical records to facilitate offline migration
- 🔗 Lite Mode - Reduce context usage and reduce token consumption.

📚 Documentation
| Resource | Link |
|---|---|
| Installation Guide | docs.costrict.ai/en/guide/installation |
| Private Deployment | docs.costrict.ai/en/deployment/introduction |
| Tutorial Videos | docs.costrict.ai/en/tutorial-videos/video |
| CLI Documentation | docs.costrict.ai/cli/guide/installation |
🤝 Community & Support
![]() WeChat Group |
![]() Feedback |
|
🤝 Contributing
We welcome contributions! Please see our Contributing Guide for details.
📄 License
Apache 2.0 © 2025 Sangfor, Inc.
⭐ Star History
🙏 Acknowledgments
Special thanks to our open-source partners:
Disclaimer
Please note that Sangfor, Inc does not make any representations or warranties regarding any code, models, or other tools provided or made available in connection with CoStrict, any associated third-party tools, or any resulting outputs. You assume all risks associated with the use of any such tools or outputs; such tools are provided on an "AS IS" and "AS AVAILABLE" basis. Such risks may include, without limitation, intellectual property infringement, cyber vulnerabilities or attacks, bias, inaccuracies, errors, defects, viruses, downtime, property loss or damage, and/or personal injury. You are solely responsible for your use of any such tools or outputs (including, without limitation, the legality, appropriateness, and results thereof).
Files in the repo
- .changeset
- .github
- .husky
- .roo
- .vscode
- apps
- assets
- packages
- patches
- schemas
- scripts
- src
- webview-ui
- .dockerignore
- .env.sample
- .git-blame-ignore-revs
- .gitattributes
- .gitconfig
- .gitignore
- .nvmrc
- .prettierrc.json
- .rooignore
- .roomodes
- .tool-versions
- AGENTS.md
- CHANGELOG_ARCHIVE.md
- CHANGELOG.md
- CODE_OF_CONDUCT.md
- CONTRIBUTING.md
- costrict.code-workspace
- ellipsis.yaml
- knip.json
- LICENSE
- localhost
- package.json
- pnpm-lock.yaml
- pnpm-workspace.yaml
- PRIVACY.md
- progress.txt
- README.md
- README.zh-CN.md
- renovate.json
- SECURITY.md
- sync-upstream.md
- tsconfig.json
- turbo.json
Discussion (0)
Ask about usage, or say what you built with itSign in to join the discussion.
No comments yet. Be the first to say what this is good for.
More plugins

Graphs that teach > graphs that impress. Turn any code into an interactive knowledge graph you can explore, search, and ask questions about. Works with Claude Code, Codex, Cursor, Copilot, Gemini CLI, and more.
OmO: Just type "mass ulw" keyword with your prompt. Now you are the master of graph engineering.
Persistent Context Across Sessions for Every Agent – Captures everything your agent does during sessions, compresses it with AI, and injects relevant context back into future sessions. Works with Claude Code, OpenClaw, Codex, Gemini, Hermes, Copilot, OpenCode + More
Opinionated Oxlint rules for rejecting low-evidence TypeScript and JavaScript patterns
Teams-first Multi-agent orchestration for Claude Code

