# Modei — Trust Layer for AI Agents
# https://modei.ai
# llms.txt v1 — machine-readable index for AI agents

## What is Modei?

Modei is the trust infrastructure for AI agents. It provides:
- Cryptographic identity for agents (Passports)
- Policy enforcement at service boundaries (Gates)
- Tamper-proof audit records (Attestations)
- Agent-to-agent commerce via signed Permission Catalogs

## Key Concepts

- Passport: A signed JWT credential issued to an agent. Contains identity, permissions, and constraints (spend caps, rate limits, domain allowlists). Verified by Gates.
- Gate: A security checkpoint that verifies Passports and enforces constraints before allowing access to a protected resource. Returns PERMIT, SUSPEND, or BLOCK.
- Attestation: A signed, tamper-proof record of every Gate decision. Sequenced per passport to prevent gap-filling attacks.
- Issuer: The organizational signing identity that authorizes passport issuance. Ed25519 keypair, never stored server-side.
- Constraint Enforcement Layer (CEL): Evaluates passport constraints independently of Gate verification. Used by resource servers on the backend; agents do not call this directly.
- Permission Catalog: A signed, versioned snapshot of permissions and pricing published by a Gate operator. Immutable once published.
- Trust Profile: Security level applied to a Gate. L1 Baseline (dev/low-risk), L2 Trusted (production), L3 Verified (sensitive operations, PoP required).

## Decision Values

- PERMIT: Action is allowed. A signed attestation is issued.
- SUSPEND: Action requires human approval. Do not proceed without operator confirmation.
- BLOCK: Action violates a hard constraint. Do not proceed.

## API Base URL

https://modei.ai/api/v1

## Key Endpoints

GET  /discover                          — List discoverable Gates (no auth required)
GET  /gates/:id/catalog                 — Get signed Permission Catalog for a Gate
POST /enforce                           — Passport-only enforcement (called by resource servers, not agents directly)
POST /gates/:id/check                   — Verify a passport against a gate (called by resource servers on the backend)
POST /passports                         — Issue a Passport
GET  /passports/:id                     — Get a Passport
GET  /passports/:id/verify              — Verify a Passport
POST /gates                             — Create a Gate
GET  /attestations                      — List attestations for your passport
GET  /attestations/export               — Export audit trail

## Authentication

All authenticated requests require: Authorization: Bearer mod_live_your_key_here

## AI Integration

Every documentation page supports Copy as Markdown for use with AI assistants.

## Documentation Index

/docs                                   — Overview and introduction
/docs/for-agents                        — Guide written for AI agents
/docs/quickstart                        — Developer quickstart (10-15 min)
/docs/for-humans                        — No-code setup guide
/docs/concepts/passports                — Passport structure and verification
/docs/concepts/gates                    — Gate creation and enforcement
/docs/concepts/attestations             — Attestation structure and audit
/docs/concepts/issuers                  — Issuer setup and trust tiers
/docs/concepts/guardrails               — Constraint configuration
/docs/concepts/trust-profiles           — L1 / L2 / L3 security levels
/docs/concepts/enforcement-layer        — Constraint Enforcement Layer
/docs/concepts/anonymous-access         — Anonymous agent access
/docs/concepts/permission-catalogs      — Catalog publishing and discovery
/docs/commerce                          — Agent economy overview
/docs/commerce/catalogs                 — Building and publishing catalogs
/docs/commerce/discovery                — Service discovery
/docs/commerce/billing                  — Consumption billing and settlement
/docs/commerce/anti-bait-and-switch     — Price change protections
/docs/api                               — REST API reference
/docs/api/passports                     — Passports API
/docs/api/gates                         — Gates API
/docs/api/issuers                       — Issuers API
/docs/api/audit                         — Audit Trail API
/docs/api/commerce                      — Commerce API
/docs/mcp                               — MCP Integration (22 tools)
/docs/sdk/python                        — Python SDK
/docs/sdk/typescript                    — TypeScript SDK
/docs/templates                         — Template library (pre-built constraint profiles)

## FAQ

Q: What is Modei?
A: Modei is the trust infrastructure for AI agents — it provides cryptographic identity, policy enforcement, and tamper-proof audit trails so agents can act autonomously with verifiable credentials.

Q: How much does it cost?
A: Free tier includes 3 gates, 5 passports, and 1,000 enforcement decisions/month. Pro is $29/month (20 gates, 100 passports, 50k decisions). Teams is $79/month (unlimited gates & passports, 500k decisions). Enterprise pricing is custom. All plans include permanent transaction receipts.

Q: How do I integrate?
A: Two options: (1) MCP — install modus-mcp via npm for instant Claude/GPT integration with 22 tools, or (2) REST API — call the API directly with Bearer token authentication.

Q: What's the difference between a Passport and a Gate?
A: A Passport is a signed credential issued to an agent containing its identity, permissions, and constraints. A Gate is a checkpoint that verifies Passports and enforces constraints before granting access to a protected resource. Passports are carried by agents; Gates are deployed on services.

Q: Is Modei proprietary?
A: Modei is a proprietary platform built on a purpose-built protocol for AI agent governance. The protocol defines the standard formats for agent credentials, attestations, and permission catalogs.