> For the complete documentation index, see [llms.txt](https://docs.collieai.io/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.collieai.io/projects-and-policies.md).

# Projects & Policies

- [Projects and policies overview](https://docs.collieai.io/projects-and-policies/overview.md): How CollieAi organizes work with projects and policies — a multi-tenant model where projects isolate workloads and reusable policies define the security rules applied to each.
- [Projects](https://docs.collieai.io/projects-and-policies/projects.md): How CollieAi projects isolate workloads — each project has its own API keys, active policy, rate limits, and data retention, so you can run multiple AI apps with separate security configs.
- [Provider tokens](https://docs.collieai.io/projects-and-policies/provider-tokens.md): How CollieAi provider tokens work — store your OpenAI and Anthropic API keys, encrypted at rest, so your app only needs a CollieAi key to reach the upstream model.
- [Dictionaries](https://docs.collieai.io/projects-and-policies/dictionaries.md): How CollieAi dictionaries work — reusable word lists for Dictionary Match rules, with system groups across 23 languages, custom uploads, and automatic word extraction from code and schemas.
- [Policies](https://docs.collieai.io/projects-and-policies/policies.md): How CollieAi policies work — a reusable collection of security rules with an enforce or monitor mode that you can share across multiple projects and switch instantly.
- [Data retention](https://docs.collieai.io/projects-and-policies/data-retention.md): How CollieAi data retention works — set per-project body and log retention to balance confidentiality with the audit trail needed for GDPR, PCI, and compliance.