Repo Standards

Repo Standards is a reusable standards system for GitHub repositories. It covers repository policy, AI-agent instructions, CI/CD, governance, security, dependency updates, release automation, deployment guidance, and migration assessment.

Current status: V1.0. The current V1.0 release is the first stable, wizard-first, modular Repo Standards release.

Start here

If you want to...	Read this
Understand the V1.0 baseline	Roadmap
Understand versioning	Versioning
Use the wizard-first flow	Wizard-first adoption
Migrate an existing repo with today's CLI	Existing repository migration
Set up a new repo with today's CLI/manual steps	New repository setup
Understand the modular model	Modular standards
Standardize AI-agent instructions	AI-agent standardization
Choose governance and community settings	Governance model
Choose a license intentionally	License selection

Product stance

Repo Standards is built first for a maintainer standardizing a set of repositories, then for other developers who want to adopt, contribute to, or extend the standard.

The V1.0 user experience is:

1. Launch the Textual migration wizard.
2. Answer governance and license questions.
3. Review language, platform, and deployment detection.
4. Confirm AI-agent cleanup and Rulesync regeneration.
5. Review CI/CD classification.
6. Apply a safe migration plan.
7. Run assessment.
8. Open a focused standards PR.

The CLI tools remain supported for scripted adoption and troubleshooting.

What Repo Standards manages

• .repo-policy.yml repository policy
• AI-agent/editor rules through Rulesync
• GitHub Actions checks and reusable workflow strategy
• branch protection recommendations
• dependency updates
• release automation
• deployment-provider guidance
• governance files such as CONTRIBUTING.md, PR templates, issue templates, SECURITY.md, and CODEOWNERS
• migration assessment and drift checks

Source-of-truth model

In downstream repositories:

Layer	Path	Purpose
Repo policy	.repo-policy.yml	Adopted profile, modules, commands, gates, governance, release, and deploy metadata
AI rule source	.rulesync/rules/*.md	Canonical AI/editor behavior
Rulesync config	rulesync.jsonc	Generated target configuration
Universal generated instructions	AGENTS.md	Generated AI-agent instructions
Editor generated outputs	.cursor/, .agents/	Generated editor-specific rules and memories

Generated AI/editor files are committed, but they are not the source of truth.

Contributing

See CONTRIBUTING.md. Contributions should preserve the phased migration approach and keep docs, templates, and AI rule source aligned.