# Certiv for developers and agents

Markdown twin of the `/developers` page: a single entry point to every
machine-readable surface Certiv publishes for AI agents and developers.

## What Certiv is

Certiv is runtime assurance for AI agents. It discovers every AI agent operating
in an organization's environment, captures each agent's reasoning and tool
calls, and enforces policy before a risky action executes. Certiv is a
governance and observability layer that sits around your agents, not a model.

## Integrate with the API

- Base URL: https://api.certiv.ai
- OpenAPI 3.1 spec: https://app.certiv.ai/openapi.json
- API reference (markdown): https://app.certiv.ai/api.md
- Authentication guide for agents: https://app.certiv.ai/auth.md

Authenticate with an organization-scoped bearer token, then call the product
API. The published endpoints are `/policies`,
`/policies/{policy_id}`, plus `/auth/login` and `/auth/refresh` for tokens.
Errors are returned as JSON, never HTML.

## Discovery and catalog surfaces

- Machine-readable app index: https://app.certiv.ai/llms.txt
- API catalog (RFC 9727 linkset): https://app.certiv.ai/.well-known/api-catalog
- AI catalog (AIR): https://app.certiv.ai/.well-known/ai-catalog.json
- AI plugin manifest: https://app.certiv.ai/.well-known/ai-plugin.json
- Agents manifest: https://app.certiv.ai/agents.json
- A2A agent card: https://app.certiv.ai/.well-known/agent-card.json

## Authentication and identity surfaces

- Protected-resource metadata (RFC 9728):
  https://app.certiv.ai/.well-known/oauth-protected-resource
- Authorization-server metadata (RFC 8414):
  https://app.certiv.ai/.well-known/oauth-authorization-server

## Skills and tools

- Product skills index:
  https://app.certiv.ai/.well-known/agent-skills/index.json
- MCP server card: https://app.certiv.ai/.well-known/mcp/server-card.json

## Company and trust

- About Certiv: https://app.certiv.ai/about.md
- Contact: https://app.certiv.ai/contact.md
- Privacy: https://app.certiv.ai/privacy.md