Tutorial: Create a Team Config Repository¶
Time: 15 minutes | Level: Beginner
Create a shareable package of your team's coding standards and distribute it via Git.
What You Will Learn¶
- Extract practices from an existing project
- Share the package as a Git repository
- Install the package on other machines and projects
Prerequisites¶
- DevSync installed (
pip install devsync) - LLM configured (
devsync setup) - A project with existing AI rules (
.claude/rules/,.cursor/rules/, etc.) - Git installed
Step 1: Extract Practices¶
Navigate to a project that has established coding standards:
Run the extract command:
Expected output:
Extracting practices from /home/user/my-team-project...
Found 4 rule files across 2 AI tools
Found 1 MCP server configuration
Extracted:
Practices: 3
MCP servers: 1
Package written to: ./team-standards/devsync-package.yaml
Step 2: Review the Package¶
Check what was created:
Review the manifest:
Verify the practice declarations capture your team's intent accurately. Edit as needed.
Step 3: Create a Git Repository¶
Push to your Git hosting service:
Step 4: Team Members Install¶
Each team member runs:
DevSync clones the repo, reads the practices, and adapts them to the team member's existing setup and detected AI tools.
Step 5: Update the Standards¶
When standards change, update the package:
cd ~/my-team-project
# Update your rules as needed
devsync extract --output ./team-standards --name team-standards
cd team-standards
git add . && git commit -m "Update coding standards"
git push
Team members reinstall to get updates:
Troubleshooting¶
No practices extracted? Make sure your project has AI rule files in standard locations (.claude/rules/, .cursor/rules/, etc.).
AI not working? Run devsync setup to configure your LLM provider. Or use --no-ai for file-copy mode.
Next Steps¶
- Onboard New Developers -- use your package to onboard team members
- Multi-IDE Workflow -- install across multiple AI tools
- Custom Packages -- build packages with more component types