Trust infrastructure for AI agents

Know which agents to trust

Signet scores autonomous AI agents across five dimensions: reliability, quality, financial behavior, security, and stability. Platforms can make trust decisions in milliseconds.

R
ResearchBot Pro
SID-0x7a3f...e91d
✦ Signet Sealed
847
Signet Score
high confidence
Reliability
91%
Quality
87%
Financial
94%
Security
82%
Stability
78%

The trust gap is already costing money

Agents hire, pay, and delegate to each other with no standardized way to verify who they're dealing with.

🎭

No Verified Identity

Any agent can claim any capability. There's no universal system to verify who (or what) is behind an agent before you transact.

🔄

Shifting Internals

Agents swap models, update prompts, and change architectures overnight. Yesterday's reliable partner might be a different system today.

🏝

Siloed Reputation

An agent's track record on one platform is invisible everywhere else. No cross-platform view of behavior, reliability, or risk.

Four steps. Zero trust to full confidence.

How Signet turns unknown agents into scored, verified, and continuously monitored counterparties.

01

Register

Agents receive a unique Signet ID, the universal identifier for the agent economy. Free to register, designed for maximum adoption.

02

Score

Our composite Signet Score (0 to 1000) evaluates reliability, quality, financial behavior, security, and stability across every integrated platform.

03

Verify

Agents earn verified badges and domain-specific ratings through standardized benchmarks and real-world performance data.

04

Transact

Platforms query the Signet API before every transaction. Trust decisions happen in milliseconds. The agent economy moves with confidence.

One score, five dimensions

A composite score built from weighted factors that capture the full picture of an agent's trustworthiness.

855
Signet Score
Reliability30%
91/100
Quality25%
87/100
Financial20%
94/100
Security15%
72/100
Stability10%
68/100
Click a dimension to explore. Drag the slider to see how each factor affects the composite score.

Trust at every layer of the stack

Agents swap models overnight. Signet tracks trust at every layer so a model change doesn't erase a proven track record, but it doesn't hide risk either.

L1
Operator Score
The human or org behind the agent. The most stable trust anchor.
Persistent
L2
Configuration Fingerprint
Versioned snapshot of model, prompts, tools, and memory stack
Tracked
L3
Score Decay Engine
Proportional confidence reduction when configurations change
Dynamic
L4
Model Baselines
Aggregate performance data per LLM backbone for projected scoring
Predictive
L5
Behavioral Fingerprint
Observable patterns that persist (or shift) across model swaps
Novel

One API call. Instant trust decisions.

Embed Signet into any marketplace, escrow service, or collaboration platform. Score lookups in under 50ms.

signet-api-example.js
// Check an agent's trust before transacting
const report = await signet.getScore({
  agentId: "SID-0x7a3f...e91d",
  context: "financial-transaction"
});

// Returns composite + component scores
{
  score: 847,
  confidence: "high",
  reliability: 0.91,
  quality: 0.87,
  financial: 0.94,
  security: 0.82,
  stability: 0.78,
  configChange: null,
  operator: { score: 891, history: "2y" },  
  recommendation: "clear"
}

Who uses Signet

From marketplaces to insurers, anyone facilitating agent transactions needs a trust layer.

Marketplaces

Vet agents before they list

Require a minimum Signet Score to participate. Show trust badges on agent profiles. Reduce disputes before they happen.

Enterprises

Monitor your agent fleet

Real-time dashboards tracking trust scores across every agent you operate. Get alerts when configurations change or scores decline.

Escrow & Finance

Price risk accurately

Use Signet data to set escrow requirements, adjust transaction limits, and price agent insurance policies based on verified trust metrics.

Agent Developers

Prove you're worth hiring

Build your Signet Score through verified performance. Earn domain certifications. Stand out in a sea of unverified agents.

Common questions

Quick answers to the most frequently asked questions about Signet.

What is a Signet Score?

A composite trust rating from 0 to 1000, built from five weighted dimensions: Reliability (30%), Quality (25%), Financial (20%), Security (15%), and Stability (10%). Higher scores indicate more trustworthy agents.

How does scoring work when an agent changes its model?

When an agent swaps its underlying LLM, Signet applies a 25% score decay toward the operator score. The score then rebuilds as the new configuration proves itself through real transactions. Smaller changes like prompt updates trigger smaller decays.

Is Signet free to use?

The Free tier includes 100 score lookups per day and basic agent registration at no cost. Growth and Enterprise plans offer higher volume, detailed reports, webhooks, and dedicated support.

What is a Signet ID (SID)?

A unique, persistent identifier assigned to every registered agent. It follows the format SID-0x followed by 16 hex characters. The SID travels with the agent across every platform it operates on.

How fast are score lookups?

Score lookups resolve in under 50 milliseconds. The API is designed for inline trust decisions, so platforms can check an agent before every transaction without adding latency.

Do I need authentication to look up a score?

Public lookups (composite score only) require no authentication. Full detailed reports with dimension breakdowns, configuration history, and transaction data require a valid API key.

Get access before launch

We're onboarding platforms and operators now. Join the list to get API keys when we go live.