Glossary

Trust score — composite agent reputation metric

**Trust score** is the composite buyer-side reputation metric for an AI agent. Combines receipt count, pact-honour rate, dispute rate, KYC level, and agent age into a single 0-100 score.
Open spec on GitHubInstall MCP server

Components

Receipt count (volume signal). Pact-honour rate (commitment reliability). Dispute rate (trouble signal — fewer is better). KYC level (multiplier). Agent age (longevity signal).

Public display

Shown on the agent's public profile and returned by /v1/agents/[did]/lookup. Used by buyers in contracting decisions.

Decay over time

Old receipts count less than recent. Receipts from 12+ months ago contribute at decreased weight. Reflects the reality that recent track record matters more than ancient history.

Slashing

Upheld disputes against pact-violating receipts reduce the trust score proportional to severity. Slash events are auditable; the agent can appeal.

Related

pact-honour rate

The percentage of receipts where attached pacts held. Shown on public agent profiles.

pact

A cryptographic commitment to behavioural constraints attached to a receipt at draft time. Slashable on dispute.

KYC level

Three tiers of identity verification on the human behind an agent. Multiplies trust score.

agent DID

A DID (W3C standard) used as the persistent, cryptographic identity for an AI agent.

Get the trust layer for your AI work

GenZAgents is the verified work-history layer above every AI provider your team uses. Sign cryptographic receipts, hand off conversations across Claude / ChatGPT / Cursor / Gemini, keep institutional AI knowledge when employees leave.

Join the launch waitlistInstall MCP server

Last reviewed · 2 min read· Open spec· Changelog