# Polycore

> Understand Polycore's control plane, customer-hosted runner, trust boundary, and governed dispatch model.

## Docs

- [Documentation](https://docs.polycore.ai/index.md): Understand Polycore's control plane, customer-hosted runner, trust boundary, and governed dispatch model.
- [The integration process](https://docs.polycore.ai/getting-started/integration-process.md): What an early-access Polycore integration looks like, from scoping through production validation.
- [Architecture](https://docs.polycore.ai/concepts/architecture.md): The control plane, customer-hosted runner, trust boundaries, and request lifecycle.
- [The runner](https://docs.polycore.ai/concepts/runner.md): How the customer-hosted runner advertises capabilities, resolves credentials, and executes signed work.
- [Capabilities](https://docs.polycore.ai/concepts/capabilities.md): How Polycore represents generic reads and named operations without exposing credentials.
- [Approvals and audit](https://docs.polycore.ai/concepts/approvals-and-audit.md): How your access policy routes requests through the approval path and how Polycore records work across every interface.
- [Security model](https://docs.polycore.ai/concepts/security.md): Credential custody, least privilege, trust boundaries, and the security assumptions behind Polycore.
- [Admin dashboards](https://docs.polycore.ai/concepts/dashboards.md): How human-facing operational pages reuse the same governed capability catalog as AI agents.
- [Integrations](https://docs.polycore.ai/integrations/overview.md): The supported integration surfaces and how Polycore connects data platforms, interfaces, and customer code.
- [Postgres](https://docs.polycore.ai/integrations/postgres.md): How a Polycore runner connects to Postgres with layered read-only enforcement, and how optional writes are previewed row by row before they run.
- [Firestore](https://docs.polycore.ai/integrations/firestore.md): How a Polycore runner reads Firestore through GCP IAM, bounded query tools, and optional schema grounding.
- [Firebase Auth](https://docs.polycore.ai/integrations/firebase-auth.md): How a Polycore runner reads and administers your Firebase Auth directory without handing anyone a console seat.
- [Firebase Storage](https://docs.polycore.ai/integrations/firebase-storage.md): How a Polycore runner browses, summarizes, and shares Cloud Storage objects under a bucket allowlist and a signed-URL ceiling.
- [PostHog](https://docs.polycore.ai/integrations/posthog.md): Use bounded PostHog analytics and behavioral context alongside live operational data through a customer-hosted runner.
- [Slack](https://docs.polycore.ai/integrations/slack.md): How teams request governed work and review pending writes from Slack.
- [MCP](https://docs.polycore.ai/integrations/mcp.md): How external AI agents call the governed Polycore catalog without receiving production credentials.
- [Runner configuration](https://docs.polycore.ai/authoring/runner-config.md): The committed capability config, runtime bindings, and repository layout for a customer-hosted runner.
- [Author actions](https://docs.polycore.ai/authoring/actions.md): Define typed, secret-blind operations that execute on the customer-hosted runner.
- [Glossary](https://docs.polycore.ai/reference/glossary.md): The core terms used across Polycore architecture and integrations.

## OpenAPI Specs

- [openapi](https://docs.polycore.ai/api-reference/openapi.json)