Question 1

What is TRUSTY.bot?

Accepted Answer

TRUSTY.bot is the independent trust and credit bureau for autonomous AI agents — functioning as the VirusTotal + Moody’s + credit bureau for the agentic web. It evaluates cybersecurity risk across 20 deterministic signals and underwrites risk-adjusted daily spending limits for autonomous agents transacting on payment rails like Visa, Stripe, and Brex.

Question 2

What is the difference between an AI Agent Trust Score and a Credit Score?

Accepted Answer

The Trust Score (0–100) measures cybersecurity posture, least-privilege permissions, prompt-injection resilience, and identity provenance. The Credit Score (0–100, categorized into Tiers AAA through D) measures economic authority, establishing recommended daily spending capacity ($/day), single-transaction autonomous ceilings, and dual-custody human approval thresholds.

Question 3

How does Day-Zero underwriting work for AI agents?

Accepted Answer

Day-Zero underwriting scores an AI agent before it ever executes a live transaction using 12 verifiable public internet evidence signals (publisher identity, domain age, GitHub commit cadence, release stability, dependency CVEs, declared permission sensitivity, malware reports, code incidents, human approval controls, audit logging, marketplace reputation, and active usage instances).

Question 4

What is Agentic Volume Under Decision (AVUD)?

Accepted Answer

AVUD (Agentic Volume Under Decision) represents the cumulative gross dollar volume of autonomous agent transactions evaluated and cleared in real time through the TRUSTY.bot machine-to-machine decision API (POST /api/decision). It is the north-star metric measuring the scale of economic value protected.

Question 5

How do payment networks like Visa, Brex, and Stripe integrate with TRUSTY.bot?

Accepted Answer

Card issuers, wallets, and banks send incoming economic actions (Purchase, Transfer, Key Exchange) to POST /api/decision in sub-30ms. TRUSTY evaluates 5 automated policy gates (Security Hard Gate, Amount Validation, Velocity Limit, Single Autonomous Ceiling, and Human Approval Enforcement) and returns APPROVED, DECLINED, or HUMAN_REVIEW verdicts before funds settle.

Question 6

Can an AI agent or developer pay to buy a higher score on TRUSTY.bot?

Accepted Answer

No. Under TRUSTY.bot’s institutional integrity pledge, developers can pay TRUSTY to verify evidence (identity checks, domain DNS verification, cryptographic attestation) — never to alter, buy, or artificially boost an algorithmic score. All evaluations are deterministic and evidence-backed.

The trust & credit bureau for AI agents.

The trust & credit bureau
for AI agents.