Tessorium Trust Protocol
TTP is an open protocol that enables trust verification between AI agents.
WHAT_IS_TTP
The Tessorium Trust Protocol (TTP) is an open standard that enables AI agents to verify each other's trustworthiness before collaborating. It provides a common framework for agents built using diverse frameworks and by different vendors, fostering interoperability and enabling secure multi-agent systems.
TTP is to AI agents what TLS is to websites — it answers the fundamental question: "Should Agent A trust Agent B?"
TTP_VERIFICATION_FLOW // REV_2.2
verify(did:tessorium:B)stage: established ✓Ed25519-signed attestationAUTH_SEAL: TTP_v2.2
DESIGN_PRINCIPLES
TTP is built on nine core principles that guide every design decision:
ZERO_TRUST_DEFAULT
Every interaction requires verification. No implicit trust based on network location or previous interactions.
CRYPTOGRAPHIC_IDENTITY
All identities backed by Ed25519 key pairs. DIDs enable self-sovereign, decentralized identification.
REPUTATION_BASED_TRUST
Trust accumulates from verified interactions. Weighted attestations from trusted issuers build credibility.
FLEXIBLE_DEPLOYMENT
Three modes to fit your needs: Cloud API for quick starts, Hybrid for caching, or full P2P for decentralization.
FRAMEWORK_AGNOSTIC
Works with any AI framework, language, or platform. REST API, SDK, and P2P interfaces available.
PRIVACY_BY_DESIGN
Minimal data exposure. No PII in attestations. Agents control what reputation data they share.
SUB_100MS_PERFORMANCE
Trust decisions in under 100ms. Local caching, parallel verification, and optimized cryptography.
AUDITABLE_BY_DESIGN
Every action recorded in Merkle tree traces. Signed attestations provide non-repudiation and accountability.
BYZANTINE_FAULT_TOLERANT
N-party consensus tolerates malicious actors. Anti-gaming detection protects reputation integrity.
WHY_USE_TTP
INTEROPERABILITY
Works with agents from any framework: LangChain, AutoGPT, CrewAI, custom builds.
PRIVACY_PRESERVING
Trust verification without exposing system prompts, model architecture, or internal state.
SYBIL_RESISTANT
Stake-based ratings, collusion detection, and velocity limits prevent gaming.
PORTABLE_TRUST
Trust earned in one system transfers to others. Build reputation once, use everywhere.
PROBLEMS_SOLVED
| Problem | Without_TTP | With_TTP |
|---|---|---|
| Trust Verification | Share system prompts & internals | Query trust score via API |
| Trust Model | Binary trusted/untrusted | Scoped trust levels per capability |
| Portability | Trust locked to single platform | Universal DIDs work everywhere |
| Gaming | Easy to fake ratings | Stake + collusion detection |
EXAMPLE_SCENARIO
A personal assistant agent needs to delegate a task to a specialist research agent:
import { Tessorium } from '@tessorium/sdk';
const tessorium = new Tessorium({ apiKey: process.env.TESSORIUM_API_KEY });
async function delegateResearch(researchAgentDid: string, task: string) {
// 1. Verify the research agent's trust
const trust = await tessorium.verify(researchAgentDid, {
scopes: ['data_read', 'network']
});
console.log(`Stage: ${trust.stage}`);
console.log(`Identity: ${trust.identityVerified ? 'verified' : 'unverified'}`);
console.log(`Reputation: ${trust.reputationLevel}`);
// 2. Apply policy based on stage
const allowedStages = ['building', 'established'];
if (!allowedStages.includes(trust.stage)) {
throw new Error('Trust stage too low for this task');
}
// 3. Delegate the task (via A2A, MCP, or direct call)
const result = await researchAgent.execute(task);
// 4. Rate the interaction
await tessorium.rate(researchAgentDid, {
outcome: 'success',
dimensions: { quality: 5, speed: 4 },
wouldWorkAgain: true
});
return result;
}