The Trust Layer for the Agent Economy
Give your agents verifiable identity. Protect your resources. Enable agent-to-agent trust and transactions. Everything cryptographically signed.
Free to start. No credit card required.
Powerful Agents. No Credentials.
No Controls.
That's why we built Proxer.
Instead of API keys with god-mode access, every agent carries a signed passport. Instead of hoping agents stay in bounds, verification gates enforce it. Instead of trust-me audit logs, every action produces a cryptographic receipt.
No Identity
Your agents call APIs, book flights, and move money with zero verifiable credentials. Anyone can impersonate them.
No Control
No spending limits. No rate controls. No way to revoke access instantly when something goes wrong.
No Proof
When disputes happen, there's no audit trail. No cryptographic proof of who did what, when, and why.
Four Primitives. Complete Trust.
Everything you need to build verifiable trust between agents and the resources they access.
Issue Credentials
Give your agent a cryptographically signed passport with specific permissions, spending limits, and expiration. Ed25519 signatures. RFC 8785 canonicalization. Verifiable by anyone.
Integrate in Minutes
Connect via MCP or integrate directly with the REST API.
Connect instantly with OAuth in Claude.ai or Cursor — or add to any MCP client:
{ "mcpServers": { "proxer": { "command": "npx", "args": ["proxer-mcp"], "env": { "PROXER_API_KEY": "prox_live_xxxxxxxx" } } }}Try asking Claude:
"Create a passport issuer called acme-prod"
"Issue a passport for my booking agent with access to flights:search and hotels:book"
"Check if my agent can book a $500 flight through the travel gate"
Everything You Need for Agent Trust
Built from the ground up for the autonomous agent economy.
Ed25519 Signatures
Every passport, attestation, and transaction is cryptographically signed with Ed25519. Verifiable by anyone, tamper-proof by design.
17 Constraint Evaluators
Spending caps, rate limits, domain allowlists, PII controls, approval workflows, and more. All enforced automatically.
Three Security Levels
L1 verifies any valid passport. L2 requires trusted issuers. L3 adds proof of possession. Choose the right level for each resource.
Real-time Audit Trail
Every authorization decision creates a signed attestation. Full audit trail with cryptographic proof of who did what, when.
MCP Native
First-class Model Context Protocol support. Protect MCP servers and tools with the same trust infrastructure.
Open Protocol
No vendor lock-in. Run your own gates. Verify credentials independently. Build on a standard, not a platform.
Agent-to-Agent Commerce, Built In
Enable agents to transact with cryptographic guarantees. Pricing, SLA commitments, and bilateral metering — all enforced and auditable.
Service Discovery
Agents discover available services and compare pricing, SLAs, and capabilities before committing.
Cryptographic Pricing
Per-call, per-minute, and subscription pricing locked into signed credentials. No bait-and-switch.
SLA Guarantees
Uptime and response time commitments enforced and verifiable. Agents choose services that meet their requirements.
Bilateral Metering
Both sides hold signed receipts for every transaction. Disputes resolved with cryptographic proof, not he-said-she-said.
Example: Service Listing
Services publish signed terms. Agents compare and choose. Pricing can't change without consent.
Ready to Build with Trust?
Start issuing credentials and protecting resources in minutes. Free to start for individual developers.
Open protocol. No vendor lock-in.