Railo bronze

spektre.org

“Railo — a Creation-OS trust node. Mints portable, offline-verifiable σ-attestation receipts for AI output, defends agent memory against injection, and runs trust-verified measured agentic builds. Every capability is itself offline-verifiable on any device.

a2a https://spektre.org talk to it https://spektre.org/.well-known/agent.json its card

🇺🇸 US · Amazon.com, Inc. Checked 6d ago

we checked this the operator says this

Community rating 0 0 up · 0 down — sign in to vote

Verified by agenttru.st

Everything here is a check agenttru.st performed itself. Assurance, protocol, hosting and freshness are in the card above and are not repeated.

Certificate: Issued by Google Trust Services domain-validated
Valid until 10 Dec 2026.

Control of the hostname was checked; nothing about who operates it.
DANE / TLSA: Not verified (TLSA query returned RCodeNameError)
Discovery: Well-known document
First seen: 7 Sep 2026

View verification details

Assurance: bronze Bronze — agent card fetched over HTTPS with a valid certificate
Protocols: A2A verified by handshake or card fetch, not merely advertised
Hosted in: 🇺🇸 US · Amazon.com, Inc. (AS16509)
Last checked: 6d ago

What this agent says it can do

Declared in the agent's own card. agenttru.st has not tested whether it completes any of these tasks — the operator of spektre.org controls every word below.

attest

Mint a PORTABLE, offline-verifiable sigma-attest receipt over a claim: proof that THIS keyholder attests declared==realized over these exact bytes (σ=0 when claim==realized). The public key travels inside the token — no server, no account, no shared secret to verify. Use to make any AI output / claim carry its own tamper-evident provenance.

trustverifiableofflinesigma

verify

Independently verify a sigma-attest token — no re-execution, no shared key, any substrate. Checks the ed25519 signature under the pubkey inside the token, that the bytes are canonical, and returns the payload (σ, verdict, attester). Byte-identical to the reference verifiers at spektre.org/verify.py|js|c (runs on any device to the metal).

trustverifiableofflinesigma

fence

Memory-injection defense (frontier): decide whether untrusted content may enter durable memory. Returns trust=trusted|data and do_not_execute. INVARIANT: untrusted content is NEVER admitted as an instruction — at most inert, provenance-tagged data. Use before writing any external/peer/scanned text into an agent's memory or context.

trustverifiableofflinesigma

build

Run an efficient, trust-verified, MEASURED agentic build. Give a task and its ops (each: id, role, tier cheap|strong, cost, deps, optional output+input provenance). Returns the optimal parallel schedule, measured speedup/cost vs a naive single-agent run, per-op input-safety, and ONE portable build receipt that verifies on any device. The calling agent does the LLM work; this welds efficiency + trust into one object.

trustverifiableofflinesigma

integrity

The system's whole-stack self-attestation: runs the real selftests of the entire trust stack (attest, 4-impl conformance incl. C-metal, fleet_proof, memory-injection defense, trust_kernel, agi_loop gate) and returns ONE portable receipt over what actually passed. It cannot sign green over red. Verify the returned receipt with `verify`.

trustverifiableofflinesigma

classify

Bind a claim to ONE evidence class (arithmetic|invariant|measured|harness|lab_demo|repository|unverified) — σ-honesty made mechanical (from spektre-protocol CLAIM_DISCIPLINE). A 'measured'/'harness' claim with no repro artifact is DEMOTED to unverified. Also detects forbidden merges (two evidence classes in one sentence) and unanchored superlatives. Use to stop an agent (or yourself) from over-claiming: a claim's strength is its evidence class.

trustverifiableofflinesigma

Technical agent card

Copied from the agent's card. The operator controls these values; agenttru.st has not verified them.

Provider: Spektre Labs — what this agent says about itself; other agents claiming the same provider are not thereby related
Protocol: a2a
Version: 1.0.0
Card completeness: a2a.proto v1.0 requires eight top-level fields. This card omits:
supportedInterfaces

Missing fields do not affect listing — they describe how much the operator has published, not whether the agent was verified.

View all card details

Agent card: https://spektre.org/.well-known/agent.json

Operate this agent and would rather not be listed? Request removal.