feat(rp-login): add walletProfile — ask who this site knows you as

[stormer78]

#144 has merged; this is rebased onto main and stands alone.

The RP side has already landed — OpenVTC/verifiable-trust-infrastructure#1212 and affinidi/affinidi-webvh-service#177 are merged. Both degrade to the entry picker until this ships, so nothing is broken in the meantime, but the dead end they were fixing stays until it does.

Why #144 was not enough

#144 made proxyLogin({}) resolve-or-bind, which is enough for an RP that only needs a token. It is not enough for the two RPs we actually have. vtc-service/admin-ui and did-hosting-ui both bind their /auth/challenge to the persona DID:

POST /auth/challenge { did: entry.principalDid }
proxyLogin({ entryId, nonce: challenge })

The DID is needed before the nonce, so it cannot come out of proxyLogin. Both pages get it today from vaultList() — which enumerates the user's whole vault, behind its own consent prompt, to answer a question about one entry. On a site with nothing bound it returns empty and the page gives up. That is the dead end #144 was meant to remove and could not reach.

The call

const { did, entryId, bound } = await wallet.walletProfile({ target });

Resolves the entry for the browser-attested origin, or raises the first-use picker and binds the answer. Mints nothing and issues no session.

entryId comes back deliberately

The objection to a page holding one was never possession — it was that learning one cost a vault-enumerating prompt. An id handed back for the site's own entry costs nothing, and passing it to proxyLogin saves the lookup this call just did.

	VTA round trips	Prompts (untrusted origin)	Vault disclosed
vaultList + proxyLogin	2	2	every matching entry
walletProfile + proxyLogin	2	1	one entry

Two prompts on a first sign-in, one after

Binding raises the picker; the sign-in raises its own consent. Folding the second into the first would mean a call that mints nothing silently pre-authorizing one that does. First contact with a site is the place to ask twice.

An already-bound lookup does not prompt at all — it discloses one DID, to the site that DID exists for, which is about to receive it inside an id_token anyway. A prompt there asks the operator to re-approve a decision they already made, which is how prompts stop being read.

principalDid is read back from the VTA rather than remembered: it is maintainer-derived, and an entry whose secret was rotated there would otherwise report a DID it no longer signs as.

Two drift traps found on the way, both now guarded

provider.ts inlined a second copy of the BridgeMethod union. Adding a method failed at its own call site with an error listing every method but the new one. It uses the type now.

content.ts cannot import (classic script), so it inlines the protocol constants under a "keep these in sync" comment with nothing enforcing it. Two invariants ride on that hand-sync:

• A method routed to a type absent from PAGE_FACING_RUNTIME_TYPES takes its origin from the page's own message body rather than the browser — and every vault entry, trust record and origin pin in this wallet is keyed on origin.
• A mistyped constant routes to no handler; sendMessage resolves undefined and the page sees a shapeless failure with nothing pointing back.

tests/page-facing-surface.test.mts reads both sources and checks them against each other. Its first assertions guard against the regexes silently matching nothing — verified by mutating a constant and confirming it fails.

RP side — merged

• fix(vtc-admin): let the wallet choose the identity, instead of reading the vault verifiable-trust-infrastructure#1212 — VTC admin console
• fix(did-hosting-ui): let the wallet choose the identity, instead of reading the vault affinidi/affinidi-webvh-service#177 — did-hosting UI

Neither wallet call is removed, so the repos could land in any order; they went first.

Pre-merge checklist

• npm run lint clean (tsc -b, not --noEmit)
• npm run build clean
• npm test — 660 tests, 0 failures (4 new)
• MV3 invariants: dist/background.js single bundle, no dynamic import(), no chrome.cookies, no static content_scripts, no cookies permission
• Module boundaries + entry points unchanged (packages/core untouched)
• R1.2 — no new outbound fetch; reuses existing offscreen handlers
• R3.7 — no matching on message text
• RUNTIME_WALLET_PROFILE is in PAGE_FACING_RUNTIME_TYPES, so its origin comes from the browser — now enforced by a test rather than by a comment
• No compatibility fold — new method, nothing widened or dual-accepted