APIR
APIRAPIR · FOR AUDITORS AND SECURITY REVIEWERS

You were handed a vendor’s word about its AI. Check the record yourself.

The questionnaire comes back with a yes in every row. The architecture diagram is a slide. Neither tells you what the agent has done.

You need something you can test without the vendor’s help and without trusting whoever hosts it.

WITHOUT A RECORD

You sign off on a description.

WHAT YOU DO
STEP 1

Check the agent

Paste its passport ID. The signature is verified against the published key and the mandate is read back. A public passport’s page recomputes the whole log book as you open it. Check an agent.
STEP 2

Ask the vendor for the export

A zip of the log book with its RFC 3161 timestamp tokens. It costs them nothing to produce.
STEP 3

Run the checker on your own machine

Open source, MIT, Python, no dependencies. Every file, every hash and link, and each chain head against its token. apir-record.
STEP 4

Read the known gaps

The specification gives every object’s exact form, the chain rules and test vectors, and lists what the record does not cover. The specification.
WHAT YOU GET
FREE

Everything on this page

Checking is free for a person, with no account and no key. That does not change.
FREE

The published keys

The keys every passport is signed with are public, so you can verify a signature in your own code. Verify a passport.
WHAT IT WILL NOT DO
  • The record shows what the agent reported, and any later change to it. It cannot show what was left out.
  • A later rewrite of anything before a timestamp would no longer match it. Entries written since the last stamp are not covered yet.
  • The mandate is the owner’s own statement.
  • APIR certifies no one. It is not an insurer or a broker, and it holds no SOC 2, ISO 27001 or penetration test report.

Start with the free step.

No card and no call. Every price is on the pricing page, and what APIR holds and does not hold is on the trust page.

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.

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.