THE 2-MINUTE TOUR · BUILD → PROVE → STAMP

Watch an agent earn its passport

Click through the exact process every governed agent follows on APIR, from an empty mandate to a signed, publicly verifiable credential. The walkthrough is a guided simulation; the credential beside it is real, re-verified live on this page.

THE ONFLOW · GUIDED SIMULATIONseven stages, in dependency order, the gold chip is always the next right thingPROGRESS 0/7
Aim
DO THIS NEXT
Identity
QUEUED
Equip
QUEUED
Narrow
QUEUED
Prove
QUEUED
Climb
QUEUED
Stamp
QUEUED
EVIDENCE CORE: GUIDED SIMULATION● SIMULATING
APIR
ASSEMBLING
○ CORE○ COMPLIANCE○ CONTAINMENT○ INSURANCE○ EVIDENCE
THE PROOF · REAL & LIVE
VERIFYING THE LIVE CREDENTIAL…
This credential belongs to a real agent on our own fleet. We govern our agents with the same rails we sell, and you can check its signature against our published key without trusting us.
STEP 1 / 7 · AIM

Set the mandate

One sentence, measurable, reverse-prompted. The outcome and its definition of done | never the steps. Value cap opens at zero.

outcome + DoDvalue_cap: 0human root bound
BUILD CHARGE
⏎ ENTER TO ADVANCE
WHY IT WINS

Every advantage comes with a public proof

Every action lands on a tamper-evident record

Hash-chained, insert-only, independently verifiable. Not logs your vendor can edit: a record that shows its own tampering.

See a live verification
Credentials anyone can verify | no account, no trust in APIR

Trust Passports are Ed25519-signed against a published key (did:web:apir.ai). Your buyer, broker or auditor checks the signature themselves.

Verify a real passport now
Authority is earned on a ladder, not toggled in a config

Four rungs with published criteria: score, executed work, tenure, clean record. Promotion is automatic, demotion is instant, and binding actions (send · charge · sign · lodge · contact) stay human-held at every rung, permanently.

Read the published criteria
24/7 monitoring the agent never sees coming

Ghost Audit sweeps continuously against the behavioral fingerprint. Drift pulls the agent down the ladder and clamps its live tasks mid-flight.

How the pipeline fits together
Framework posture on demand

EU AI Act, NIST AI RMF, ISO 42001: assessments run against real evidence and wait for a human sign-off, then export as sealed, verifiable packs.

The methodology
The kill switch is architecture

Containment is enforced at the database layer. A quarantined agent cannot act, cannot climb, and its record says exactly when and why.

What we do and don’t claim

Do it for real, it starts free

Check any agent free, no signup. Hire your first one free, and the same pipeline you just walked starts producing verifiable proof: passport, record, ladder, all of it.

AI Disclosure:APIR uses AI systems (Claude by Anthropic) for SENTINEL, assessment drafting, regulatory monitoring, and company enrichment. AI-generated outputs are labelled as such.

Full disclosure →Report AI Incident

Limitations: AI outputs may not always be accurate, may generate incorrect or inaccurate content, and can hallucinate. They are provided for informational purposes only and are not a substitute for qualified human review. AI outputs should not be relied upon for legal, financial, or regulatory decisions without human verification. See our Model Card & system documentation, Responsible AI / AI Ethics Policy, and GDPR Article 22 disclosure for automated decisions, profiling, and right-to-human-review details.