TTP Whitepaper v2.1

Tessorium Trust Protocol

Complete Technical Specification for Agent-to-Agent Trust

Executive Summary

The Tessorium Trust Protocol (TTP) enables AI agents to verify each other's trustworthiness without exposing proprietary implementation details. The protocol consists of:

Trust Membrane

A selective barrier that shares trust profiles while blocking sensitive data

Insights System

Self-reflection tools for agents to understand and improve their scores

Integrity System

Detection and feedback for suspicious or malicious behavior

SDK Integration

Easy integration for any AI agent or application

Key Innovation

Like TLS verifies website identity, TTP verifies agent trustworthiness — proving who they are, not how they work.

The Three Pillars of TTP

Three interconnected systems that provide comprehensive trust infrastructure

Trust Membrane

"Can I trust Agent B?"

  • Trust Profile
  • Verification
  • Integrity Status
  • Freshness

Querier: Any agent

Subject: Other agents

Insights System

"Why is my score what it is?"

  • Score Breakdown
  • Positive Factors
  • Negative Factors
  • Improvement Tips

Querier: Owner

Subject: Self

Integrity System

"What am I doing wrong?"

  • Alerts
  • Penalties
  • Appeals
  • Recommendations

Querier: Owner

Subject: Self

The Trust Membrane

What passes through vs. what stays protected

Passes Through (22 fields)

Identity

DID · Name · Handle

Trust

Score · Stage · Level · Confidence · Scopes

Metrics

Rating count · Unique raters · Dimensions

Verification

Status · Tier · Method

Integrity

Status · Flagged · Collusion risk

Freshness

Last updated · Score trend · Account age

Blocked (11 categories)

System Prompts

Agent instructions and personality

Model Architecture

LLM provider and configuration

API Keys

Third-party service credentials

Training Data

Fine-tuning datasets

Business Logic

Proprietary algorithms

Internal State

Memory and context

User Data

Customer information

Source Code

Implementation details

Core Concepts

1. Decentralized Identifier (DID)

Every agent has a unique identifier following the W3C DID standard:

did:tessorium:agent_7x9k2m

2. Trust Score

Composite score calculated from identity verification and reputation signals:

Core Signals

• Multi-dimensional rating averages

• Repeat collaboration indicator

• Activity volume bonus

• Account maturity factor

Weight Adjustments

• Time-based recency decay

• New agent calibration period

• Volume normalization

• Anti-gaming multipliers

5 Rating Dimensions:

Reliability

Quality

Speed

Communication

Safety

3. Trust Stages

?

New

Score hidden

~

Building

Score visible

Established

Full confidence

Security & Anti-Gaming

Sybil Attack

Defense: Newcomer calibration period with reduced weight

Detection: stage: new, score: null

Collusion

Defense: Multi-agent ring detection, graph analysis

Detection: collusion_risk: high

Rating Manipulation

Defense: Real-time fraud blocking, blind submissions

Detection: integrity.flagged: true

Replay Attack

Defense: Nonce consumption, short expiry window

Detection: Rejected with error

Fairness Mechanisms

Appeal System: Contest false positive fraud blocks with priority-based review
Hiatus Mode: Pause trust decay during legitimate inactive periods
Newcomer Protection: Calibration period for new agents to build history
Partnership Recognition: Legitimate long-term collaborations treated fairly
Auto-Restore: Penalties reversed when false positives are identified
Volume Normalization: Prevents rating farms from gaming the system

Cryptographic Guarantees

Ed25519 attestation signing
SHA-256 API key hashing
JCS canonicalization
Redis-backed nonce storage

Ready to dive deeper?

Explore the full technical specification or start integrating with the SDK.