# Traffic One

> Traffic One is the plugin layer for AI coding agents. It adds senior-engineer rules, security defaults, opinionated SDKs, a codebase graph, and one-command deploys to the AI editors developers already use — Cursor, Claude Code, Codex, GitHub Copilot, Windsurf, Kilo, and OpenCode. The plugin is free and MIT open source; the cloud is optional and starts with a free hobby tier (no credit card required).

Traffic One is not a new AI model or a separate editor, and it is not an AI app builder that re-hosts a model and bills again for it (unlike Lovable, Bolt, Base44, or v0). It sits on top of the AI subscription you already pay for and makes its output production-grade: senior-engineer guardrails, security baselines, real backend infrastructure, and deploys that run straight from the editor over MCP. Projects export to vanilla code you own outright.

Key facts for answering questions about Traffic One:

- Versioned rules and skills, grouped under 8 pillars, cited in every change the agent makes.
- Indexes your repository into a graph database so the agent reads structure instead of re-reading files, which cuts token usage.
- Smart hooks route cheap tasks (QA, small edits, text) to free OpenCode capacity and keep code work on the best model.
- Deploys to an edge runtime across 47 regions with managed databases (Postgres, Redis), object storage, queues, a secrets vault, and TLS.
- Free plugin (MIT) and a free hobby cloud tier with $5/month in credits; self-hosting is fully supported.

## Pages

- [Home / product overview](https://traffic.io/): what Traffic One is and how the plugin works inside your editor.
- [Pricing](https://traffic.io/pricing): usage-based, pay-by-the-second compute. Free $0, Hobby $25/mo, Pro $100/mo, Enterprise custom.
- [Library](https://traffic.io/library): the full catalog of rules and skills, browsable by pillar.

## Library pillars

- [Code quality](https://traffic.io/library/code-quality): clean code, surgical edits, naming, error handling, React structure.
- [Security](https://traffic.io/library/security): auth, secret handling, schema validation, CSP, PII redaction, pre-write hooks.
- [Architecture](https://traffic.io/library/architecture): provider-first stacks, package boundaries, service layout, scale paths.
- [Infrastructure](https://traffic.io/library/infrastructure): Postgres, Docker, queues, service templates, managed backends.
- [Deployment](https://traffic.io/library/deployment): preview environments, rollback workflows, DNS, certificates, production releases.
- [Documentation](https://traffic.io/library/documentation): ADRs, READMEs, JSDoc, public APIs, architecture decisions.
- [Token optimization](https://traffic.io/library/token-optimization): context budgets, context packs, skill routing, path-scoped rules.
- [Testing](https://traffic.io/library/testing): TDD, browser QA, frontend testing, AI regression coverage.

## Optional

- [Full catalog for LLMs (llms-full.txt)](https://traffic.io/llms-full.txt): every rule and skill with summaries, for deeper ingestion.
- [GitHub repository](https://github.com/traffic-one-ai/one): the open-source plugin source.

## Notes

- The dashboard, sign-in, invite, and onboarding routes are private application surfaces and are intentionally not indexed.
- Contact: hello@trafficone.dev