# Keycard — Identity & Access Management for AI Agents

# About Keycard
Keycard controls what AI agents can access, what they can do with it, and logs every authorization decision with full attribution. Access is scoped to the task and enforced by policy. Keycard is identity and access management (IAM) infrastructure for the agent-native world, providing secure identity, authorization, and governance for AI agents and MCP (Model Context Protocol) platforms.

# Company Information
- Name: Keycard
- Website: https://www.keycard.ai
- Tagline: "Runtime auth for agents"
- Mission: Unlock the power of AI agents by giving developers and enterprises the foundations they need to build and adopt trusted agentic applications at scale.

# Core Product
Keycard provides deterministic guardrails to the non-deterministic world of AI agents by evolving the way applications, agents, and machines identify and authorize one another. The core of Keycard is a secure token service that supports federation and brokering of credentials across applications, agents, and services. Keycard also offers an optional gateway, and an integration toolkit (SDK) is in development so Keycard can integrate with any existing gateway.

Key features:
- Ephemeral, identity-bound tokens with support for mixed delegation chains
- Task-scoped policy enforcement
- Dynamic provisioning and revocation of access at runtime
- Federated identity across systems and companies
- Every authorization decision and credential request recorded and attributed, denials included
- Contextual audit logs for complete visibility
- MCP-native integration

# Funding & Investors
- Total Funding: $38M combined inception round
- Lead Investors: Andreessen Horowitz (a16z), Acrew Capital, Boldstart Ventures
- Participating Investors: Mantis VC, Tapestry Ventures, Essence Ventures, Exceptional Capital, Modern Technical Fund, Vermilion Cliffs Ventures

Individual Investors:
- Matias Woloski, CTO, Auth0
- Karl McGuinness, Chief Product Architect, Okta
- Ian Andrews, Former CRO, Groq
- Ryan Carlson, President, Chainguard
- Emilio Escobar, CISO, Datadog
- Dane Knecht, CTO, Cloudflare
- Angie Lai, Head of Security Engineering, Anthropic
- Martin Gontovnikas, Early growth, Auth0, CMO, Vercel
- Ian Swanson, CEO, Protect AI

# Key Pages & Content

## Home (/)
Page Title: Keycard — Runtime auth for agents
Page Description: Runtime authorization for AI agents. Give each agent its own identity, authority to act, and access to resources under your policy.

Main sections:

### Hero
Title: Runtime auth for agents
Subtitle: Give each agent its own identity, authority to act, and access to resources under your policy.
Calls to action: Book a demo (contact sales), Read the docs.

### Customer story
A quote from Chime's Principal Product Manager, Generative AI, on running agents with Keycard.

### The challenge
Title: The better your agents get, the more access they need.
Three approaches that don't hold up:
- Put a human in front of every action? Manual approvals don't scale.
- Let an agent inherit a human or workload identity? There's no way to tell who took what action.
- Provision for everything it might need? That's also everything it could do.

### How Keycard works
Subtitle: Keycard verifies each request, evaluates it against policy, and makes authorization decisions for your agents and whatever they need to access.
Four steps:
1. Capture the full context. Keycard verifies the agent making the request and where it's running. When it acts for a user, the request also carries that user's identity. The full picture is used to make authorization decisions at runtime.
2. Make resources available. Connect a resource once and make it available to whoever needs it. Resources can be data stores, MCP servers, APIs, or agents. Configure the resource's credential provider and Keycard will handle all authorization requests against it.
3. Grant access dynamically. At runtime, each request is evaluated against policy. If approved, Keycard's secure token server will issue a short-lived authorization token scoped to the task at hand. If denied, no credentials are created but the decision is logged for visibility.
4. See real-time activity. Inspect sessions by application, user, or resource and export audit logs to your SIEM. Each authorization event includes full attribution, including denials. Revoke a credential instantly and subsequent events are denied wherever it applies.

### Policy as code
Title: Write the rules. Apply them at runtime.
Define access in Cedar and manage changes using Terraform. Validate and test new policies, then activate the version you want Keycard to use for credential requests. The panel shows the same policy as Cedar and as a visual rule.

### Enforcement points
Title: Integrate Keycard where it's needed
- In your application: add authentication, authorization, and delegated API access to any agent or MCP server with a few lines of code.
- Through a gateway: a gateway brokers authentication and authorization across agents, clients, and resources, with no code changes.
- In your infrastructure: a sidecar proxy intercepts the agent's outbound traffic and swaps placeholder secrets for real, user-scoped tokens from Keycard. The agent never holds a real credential.

### Standards
Title: Built on open standards. Helping write the new ones.
Keycard speaks the protocols your stack already uses, ships drop-in SDKs for the services you build, and federates the identity providers you run today. Policy is written in Cedar and applied through Terraform, the same way you manage the rest of your infrastructure.
SOC 2 Type II (Trust Center: https://trust.keycard.ai/).
- Identity: federate users and workloads with your existing identity providers.
- Build: SDKs for Python, TypeScript, Go, and Ruby.
- Policy: define access in Cedar. Configure Keycard resources with Terraform.
- Audit export: export events to any SIEM that reads OCSF, or send them to S3.
Agent Baseline (white paper, https://agentbaseline.org/): a shared framework for agent security, developed with Docker and Snyk, covering six outcomes and 35 controls.

### Testimonials
Title: What practitioners are running into.
Quotes from security practitioners on the Insecure Agents podcast (https://www.insecureagents.com/).

### FAQ (/#faq)
See the FAQ section below.

### Final call to action
Title: Give every agent a Keycard
Book a demo.

## About (/about)
Page Title: About — Keycard
Page Description: Keycard makes AI agents safe, trustworthy, and production-ready.

### Hero Section
Title: Keycard makes AI agents safe, trustworthy, and production-ready
Subtitle: Our mission is to reimagine how developers connect people, agents, and services across networks, clouds, and applications.

### Mission Cards
Three mission pillars:
1. Unblock Critical AI Projects
   Empower teams to move agent-driven workloads into production — without compromising control.

2. Democratize Agents and Tools
   Make it easy to securely connect agents to real services, across any cloud and all teams.

3. Build for the Agent Native Future
   Ship the infrastructure that makes autonomous systems trustworthy — identity, policy, and audit from day one.

### Our Investors Section
Institutional: a16z, Acrew, boldstart, mtf.vc, Essence, Mantis, Exceptional Capital, Vermilion Cliffs, Tapestry VC.
Individual investors listed above.

### Founders Section
- Matthew Creager, Product & Marketing
- Jared Hanson, Technology
- Ian Livingstone, CEO

### Open Positions CTA
Links to /careers for current job openings.

## Pricing (/pricing)
Page Title: Pricing — Keycard
Three plans, billed on one metric: transactions. A transaction is recorded each time Keycard is involved in a request: issuing a credential, validating an access request, or exchanging a credential. Every plan includes unlimited users, agents and apps; RBAC, ABAC and ReBAC; real-time telemetry; and every supported protocol.
- Starter: Free, 5,000 transactions/mo (hard cap). App Binding + Dependencies policy, 7-day telemetry retention, community support.
- Team: $500/mo with 100,000 transactions included, billed monthly with no commitment; $1 per 1,000 additional transactions. Everything in Starter, plus SSO for organization users, Zone Policy, 90-day telemetry retention, full usage data & insights, and email support.
- Enterprise: Contact sales for custom transaction volume, on an annual commitment. Everything in Team, plus Org Policy and device-based policy, SCIM, Active Directory and LDAP provisioning, custom roles, continuous export (OCSF), dedicated, BYOC or on-prem deployment, private networking, customer-managed KMS, 180-day telemetry retention (extendable), a 99.95% uptime SLA, and 24/7 support with a dedicated CSM.

## Contact sales (/contact)
Page Title: Contact sales — Keycard
Page Description: Tell the Keycard team what you're building and what your agents need access to, and we'll follow up by email to set up a call.
Contact Sales form (name, work email, title, company, message). Submissions go to the sales team; a person follows up by email.

## Careers (/careers)
Page Title: Careers — Keycard
Page Description: Join Keycard. Build identity infrastructure to power the next generation of AI.

### Hero
Title: Join Keycard
Subtitle: Build identity infrastructure for agents

### Open positions
All positions are remote. Not seeing your exact role? We're always looking for good people. Reach out: hello@keycard.ai

### Why work with us
Backed by a16z, Boldstart, and Acrew. Competitive comp + equity, full benefits, remote-first.
1. The problem is open — Agents can't go to production without identity and access control. Nobody has shipped the answer yet. You'd be building it.
2. Give every team agents — Most teams can't connect agents to real services because the security story isn't there. Your work removes that barrier.
3. Define the category — Agent-native infrastructure barely exists. We're writing the spec, building the product, and talking to the people who need it.

## FAQ (/#faq)
The FAQ is a section of the homepage (https://www.keycard.ai/#faq); /faq redirects there. The same six questions are published as FAQPage structured data on the page.

Q: What is Keycard and what does it do?
A: Keycard is runtime authorization infrastructure for AI agents. It gives applications their own identity, evaluates credential requests against policy, and issues credentials when access is allowed. Register applications and resources once, then reuse them as you build.

Q: Who is Keycard for?
A: Developers building agents and internal tools, alongside platform, security, and AI operations teams bringing them into production. Use Keycard with your agents, MCP servers, APIs, and other services.

Q: When should a company use Keycard?
A: When agents need to act on production systems at scale. If you’re managing dozens of MCP servers, using multiple human and workload identity providers, or considering refactoring existing applications one-by-one, that’s a sign you need Keycard.

Q: How does Keycard work with my existing stack?
A: Keycard federates the identity provider you already run, whether that’s Okta, Entra, Google, or Auth0, and adds policy and token issuance on top rather than replacing it. SDKs for Python, TypeScript, Ruby, and Go cover services your developers build. Policy is defined in Cedar and ships as code through Terraform. Supported audit events export to your Amazon S3 bucket for ingestion by compatible SIEM tools. If you already run an AI gateway, talk to us about how it can act as the enforcement point for authorization decisions.

Q: What is runtime authorization for AI agents?
A: Runtime authorization for AI agents means the access decision is made when the agent acts, not when access was set up or when the agent was deployed. Each request carries context on which agent is asking, where it’s running, who it’s acting for, and what it wants to reach. Policy is evaluated against the full composite identity, and if approved, a credential is issued for the approved access.

The alternatives are to let the agent inherit a human identity or a shared service account with standing permissions; to pre-authorize the agent with its own fixed set of permissions, sized in advance to cover everything it might need; or to gate progressive privilege by human-in-the-loop approvals. Each gives something up. Inherited access is sized to someone else’s job and attributed to them rather than to the agent. A pre-authorized grant is written before anyone knows what the agent will do, so it covers everything it might do. Approvals fit a small set of sensitive operations, but they don’t scale to every request.

Q: How is Keycard’s approach to agent access different from other solutions?
A: Most solutions inspect traffic after a credential already exists. Keycard evaluates each request against policy and issues credentials only if access is approved. If the request is denied, Keycard issues no new credential. Keycard records policy decisions and credential requests in audit logs. Keycard mints the tokens, but your traffic doesn’t have to go through it. With the SDK in your app or a proxy in your infrastructure, your agent talks to the resource directly. Calls pass through Keycard only if you route them through its gateway, which is optional.

## Trust Center (https://trust.keycard.ai/)
Security posture, compliance (SOC 2 Type II), and trust-related information, powered by SafeBase.

## Blog (/blog)
Page Title: Blog — Keycard
Page Description: Insights on agent security, identity, and the future of governed AI.
Product updates, announcements, and thought leadership from the Keycard team.

## Brand (/brand)
Page Title: Brand — Keycard
Page Description: Official Keycard brand assets: logos, wordmarks, and press kit downloads, plus team bios for press and partners.

Includes:
- Logo & combinations (default, green, background variations, glowing effects, flowing effects)
- All logos available in PNG and SVG formats
- Team photos and bios

## Privacy (/privacy)
Privacy policy and data processing information (GDPR & CCPA compliant).

## Cookie Policy (/cookie-policy)
How Keycard uses cookies and how visitors manage consent preferences.

## Platform (/platform)
The control plane for autonomous agents. (Coming soon)

# Social Media & Links
- GitHub: https://github.com/keycardai
- X (Twitter): https://x.com/keycardai
- LinkedIn: https://www.linkedin.com/company/keycard-ai

# Key Concepts

## Auth for agents
From "For every action, the right access: defining auth for agents" (Ian Livingstone, October 8, 2026): https://www.keycard.ai/blog/defining-auth-for-agents/

Auth for agents comes down to two things:
1. Deciding an agent's access from the full context of the task: who it's acting for, what the task requires, and what the organization allows.
2. Enforcing access as the work happens across every agent and every system it touches, and bringing people back in when an agent needs more.

The scope of access is the intersection of what the agent can do with the authority it has and what the task needs, all within the bounds of what the organization allows.

Most of that information already exists around the agent. The prompt or trigger says who it's acting for and what it's meant to do. All of it disappears once the work becomes an API call with a token attached. We call that loss the context gap, and closing it is the core of auth for agents.

## AI Agent Identity
The challenge: Traditional IAM systems were designed for static, human-driven interactions, not the dynamic, ephemeral, and autonomous world of AI agents. Agents require:
- Ephemeral identity that can be created and destroyed quickly
- Federated identity across systems and companies
- Attestable identity bound to device or cloud runtime
- Traceable delegations back to original users or companies

## MCP (Model Context Protocol)
Keycard is MCP-native, integrating directly with MCP platforms to provide identity and access management for multi-agent systems.

## Agent-Native Era
We're entering a new era where software is dynamic, with AI agents working on behalf of people and businesses. This requires new identity and access primitives.

## Keycard's Approach
Keycard provides deterministic guardrails to the non-deterministic world of AI agents by evolving the way applications, agents, and machines identify and authorize one another. The core of Keycard is a secure token service that supports federation and brokering of credentials across applications, agents, and services. Keycard also offers an optional gateway, and an integration toolkit (SDK) is in development so Keycard can integrate with any existing gateway.

# Use Cases
- Extend copilots & adopt agents: Adopt internal or external agents and expose tools, APIs, and services securely
- Create trusted agents & tools: Build secure, identity-aware agentic applications using Keycard's SDKs
- Build multi-agent systems: Coordinate agents, tools, and data across teams, networks, and companies
- Build next-generation products: Evolve products into agents or safely connect agents to platforms

# Early Access
Keycard is currently available in early access. Visitors request access through the "Request access" button in the site header, or book a demo with sales.

# Contact
- Email: hello@keycard.ai
- Website: https://www.keycard.ai

# Technical Details
- Base URL: https://www.keycard.ai
- Standards: OAuth 2.1, MCP, A2A
- Framework: Astro
- Trust Center: https://trust.keycard.ai/

# Site Structure
- Home: /
- About: /about
- Pricing: /pricing
- Keycard for Code: /keycard-for-code
- Careers: /careers
- Events: /events
- FAQ: /#faq (a section of the homepage)
- Blog: /blog
- Blog posts: /blog/[slug]
- Brand: /brand
- Privacy: /privacy
- Cookie Policy: /cookie-policy
- Platform: /platform (coming soon)

# Last Updated
October 2026