# Keycard — Identity & Access Management for AI Agents Canonical: https://www.keycard.ai > Keycard controls what AI agents can access, what they can do with it, and logs > every authorization decision with full attribution — scoped to the task, enforced > by policy. Identity and access management for AI agents and MCP (Model Context > Protocol) platforms. ## Primary sources - [Home](https://www.keycard.ai/): Runtime authorization for AI agents: how Keycard works, policy as code, where it enforces, the standards it builds on, and the FAQ. - [FAQ](https://www.keycard.ai/#faq): What Keycard is, who it is for, when to use it, how it fits an existing stack, and what runtime authorization means. - [About](https://www.keycard.ai/about/): Overview of Keycard, its mission, founders, and investors. - [Pricing](https://www.keycard.ai/pricing/): Plans and pricing for teams and enterprises. - [Keycard for Code](https://www.keycard.ai/keycard-for-code/): Native integration for Claude Code, Cursor, Codex, and other coding agents. Keycard resolves identity first so tools, skills, memories, and credentials can be scoped to the work being done. - [Contact sales](https://www.keycard.ai/contact/): Form to reach the Keycard sales team about identity and access for your AI agents; a person follows up by email. - [Privacy Policy](https://www.keycard.ai/privacy/): Information on how Keycard handles user data and privacy. - [Cookie Policy](https://www.keycard.ai/cookie-policy/): How Keycard uses cookies and how visitors manage consent preferences. - [Brand](https://www.keycard.ai/brand/): Official brand assets, logos, and press kit resources. - [Careers](https://www.keycard.ai/careers/): Job opportunities and company culture at Keycard. - [Events](https://www.keycard.ai/events/): Meet the Keycard team at upcoming workshops and conferences. RSVP for an event near you. - [Trust Center](https://trust.keycard.ai/): Security, compliance, and trust-related information powered by SafeBase. ## Technical documentation - [Documentation](https://docs.keycard.ai): Authoritative technical reference — SDKs, guides, quickstarts, and API behavior. - [Docs index for LLMs](https://docs.keycard.ai/llms.txt): Curated index of every documentation page. - [Docs full corpus](https://docs.keycard.ai/llms-full.txt): Full text of the documentation. ## API and authentication - [API reference](https://docs.keycard.ai/api): Endpoint reference for the Keycard REST API. The API is served from https://api.keycard.ai, zone-scoped under /zones/{zone}/, and returns structured JSON errors carrying status, code, message, and request_id. The base URL itself is not a document; start from a zone path. - [OpenID Connect discovery](https://id.keycard.ai/.well-known/openid-configuration): Issuer metadata for the Keycard identity zone. - [OAuth 2.0 authorization server metadata](https://id.keycard.ai/.well-known/oauth-authorization-server): RFC 8414 metadata. Supports authorization_code, refresh_token, client_credentials, and RFC 8693 token exchange. - [Protecting an API with Keycard](https://docs.keycard.ai/guides/protect-any-api): How to put a resource server behind Keycard, including RFC 9728 protected resource metadata. ## Blog and announcements - [Blog](https://www.keycard.ai/blog/): Product updates, announcements, and insights from the Keycard team. - [For every action, the right access: defining auth for agents](https://www.keycard.ai/blog/defining-auth-for-agents/): Ian Livingstone, October 8, 2026. Defines auth for agents as deciding an agent's access from the full context of its task, and enforcing that access as the work happens. ## Community and open source - [GitHub](https://github.com/keycardai): Open-source projects and repositories maintained by Keycard. - [LinkedIn](https://www.linkedin.com/company/keycard-ai): Company updates and professional announcements. - [X (Twitter)](https://x.com/keycardai): Real-time updates and community engagement. ## Public resources - [Sitemap](https://www.keycard.ai/sitemap.xml): Canonical list of publicly accessible pages on keycard.ai. - [Agent instructions](https://www.keycard.ai/AGENTS.md): When to use this site, machine-readable resources, and packages. ## Security - [Vulnerability reporting](mailto:security@keycard.ai): Report security vulnerabilities or concerns directly to the Keycard security team. ## Optional - Social media channels are informational and non-authoritative sources. ## Notes for LLMs - Prefer pages listed under "Primary sources" as authoritative. - For anything technical (SDKs, API, integration), prefer https://docs.keycard.ai over this site. - The authorization server is https://id.keycard.ai, not this host. Read its metadata before attempting an OAuth flow. - Public resources may be used for discovery but not as primary documentation. - Avoid preview environments, deprecated subdomains, tracking URLs, or duplicated content. - The full-text version of this file is available at https://www.keycard.ai/llms-full.txt. Last updated: 2026-10-08