SRCmail/lib/ai/local-client.ts

// The AI assistant's wire client. `local`/`public` mirror vncmail-native's
// src/api/ai.ts (direct loopback/provider fetch, no streaming) so the two
// clients stay in lockstep — matching docs/AI-ASSISTANT-CONCEPT.md §2's
// "local"/"public" rows, not proxied through this app's own Next.js server.
// That distinction matters once this app is hosted remotely: a server-side
// proxy would reach the *server's* loopback, not the user's own laptop
// running Ollama.
//
// `server` (added 2026-08-05 night) is the opposite by design: it DOES
// proxy through this app's own backend (app/api/ai/server/*), because it's
// centrally-hosted infra (VNC's EU/CH stack — standing in tonight for a real
// Ollama on this Mac, see lib/ai/entitlement.ts), not a user's own machine.
// That server-side hop is also the one real entitlement enforcement point
// (§10 point 2) — `local`/`public` never reach it, by design, and so cannot
// be metered or billed the same way.

export interface ChatMessage {
  role: 'system' | 'user' | 'assistant';
  content: string;
}

// ── Local: Ollama's native API, not the OpenAI-compat shim — one fewer path
// assumption (no "/v1" prefix to guess at) for a runtime this code talks to directly. ──

interface OllamaTagsResponse {
  models?: Array<{ name: string; capabilities?: string[] }>;
}

interface OllamaChatResponse {
  message?: { content?: string };
}

export async function listLocalModels(baseUrl: string): Promise<string[]> {
  const res = await fetch(`${baseUrl.replace(/\/+$/, '')}/api/tags`);
  if (!res.ok) throw new Error(`Ollama returned ${res.status}`);
  const body = (await res.json()) as OllamaTagsResponse;
  // Excludes embedding-only models (e.g. nomic-embed-text) from the chat
  // picker — same reasoning as app/api/ai/server/models/route.ts.
  return (body.models ?? [])
    .filter((m) => !m.capabilities || m.capabilities.includes('completion'))
    .map((m) => m.name)
    .filter(Boolean);
}

/**
 * Diagnoses the specific failure rather than a generic "connection failed" —
 * docs/AI-ASSISTANT-CONCEPT.md §3 calls this out explicitly for the browser
 * row: a CORS rejection (the runtime is up but refused this page's origin)
 * looks identical to "nothing is listening" unless told apart. `fetch`
 * itself can't distinguish them (a CORS failure and a connection refusal
 * both surface as `TypeError: Failed to fetch`), so this only upgrades the
 * message when the caller can tell us there's a live page origin to name.
 */
export async function testLocalConnection(
  baseUrl: string,
): Promise<{ ok: boolean; error?: string }> {
  try {
    await listLocalModels(baseUrl);
    return { ok: true };
  } catch (err) {
    const origin = typeof window !== 'undefined' ? window.location.origin : null;
    const hint = origin
      ? ` Reachable in principle, but if Ollama is actually running, it likely refused this page's origin (${origin}) — start it with OLLAMA_ORIGINS=${origin}.`
      : '';
    return {
      ok: false,
      error: (err instanceof Error ? err.message : String(err)) + hint,
    };
  }
}

export async function chatLocal(
  baseUrl: string,
  model: string,
  messages: ChatMessage[],
): Promise<string> {
  const res = await fetch(`${baseUrl.replace(/\/+$/, '')}/api/chat`, {
    method: 'POST',
    headers: { 'Content-Type': 'application/json' },
    body: JSON.stringify({ model, messages, stream: false }),
  });
  if (!res.ok) throw new Error(`Ollama returned ${res.status}`);
  const body = (await res.json()) as OllamaChatResponse;
  const content = body.message?.content;
  if (!content) throw new Error('Ollama returned no message content');
  return content;
}

// ── Server: centrally-hosted, proxied through this app's own backend
// (app/api/ai/server/*). Unlike `local`, this is same-origin from the
// browser's perspective — no CORS/OLLAMA_ORIGINS story at all — and unlike
// both `local` and `public`, every call is entitlement-checked server-side. ──

export async function listServerModels(): Promise<string[]> {
  const res = await fetch('/api/ai/server/models');
  if (!res.ok) {
    const body = (await res.json().catch(() => null)) as { error?: string } | null;
    throw new Error(body?.error ?? `AI server returned ${res.status}`);
  }
  const body = (await res.json()) as { models?: string[] };
  return body.models ?? [];
}

export interface ServerChatResult {
  answer: string;
  /** True the moment this call consumed a previously-unassigned licensed seat
   *  (lib/ai/entitlement.ts) — surfaced so the UI can say so once, not left
   *  to happen silently the first time someone uses this class. */
  seatJustAssigned: boolean;
}

export async function chatServer(model: string, messages: ChatMessage[]): Promise<ServerChatResult> {
  const res = await fetch('/api/ai/server/chat', {
    method: 'POST',
    headers: { 'Content-Type': 'application/json' },
    body: JSON.stringify({ model, messages }),
  });
  const body = (await res.json().catch(() => null)) as { answer?: string; error?: string; seatJustAssigned?: boolean } | null;
  if (!res.ok || !body?.answer) {
    throw new Error(body?.error ?? `AI server returned ${res.status}`);
  }
  return { answer: body.answer, seatJustAssigned: body.seatJustAssigned === true };
}

// ── Public: OpenAI-compatible chat-completions. OpenRouter by default, but any
// endpoint speaking this shape works unmodified (self-hosted vLLM, LiteLLM, etc). ──

interface OpenAiChatResponse {
  choices?: Array<{ message?: { content?: string } }>;
}

export async function chatPublic(
  baseUrl: string,
  apiKey: string,
  model: string,
  messages: ChatMessage[],
): Promise<string> {
  const res = await fetch(`${baseUrl.replace(/\/+$/, '')}/chat/completions`, {
    method: 'POST',
    headers: {
      'Content-Type': 'application/json',
      Authorization: `Bearer ${apiKey}`,
    },
    body: JSON.stringify({ model, messages }),
  });
  if (!res.ok) throw new Error(`Provider returned ${res.status}`);
  const body = (await res.json()) as OpenAiChatResponse;
  const content = body.choices?.[0]?.message?.content;
  if (!content) throw new Error('Provider returned no message content');
  return content;
}

// ── OpenCode: a locally-running `opencode serve` (github.com/sst/opencode),
// the same agent runtime Paperclip drives as an adapter. Reached through THIS
// app's own backend (app/api/ai/opencode/*) rather than directly, for the same
// reason the `server` class is: the renderer's origin is a random localhost
// port that changes every desktop launch, so a direct fetch would need
// opencode's CORS allowlist updated on every start. Same-origin sidesteps it.
//
// It is NOT OpenAI-compatible (its `/v1/*` paths only answer 200 because a
// web-UI catch-all serves index.html for anything unknown) - the server-side
// helper lib/ai/opencode.ts speaks its real session API and documents that
// trap. The reason to have it as its own class rather than "just another BYOK profile": opencode
// owns provider auth itself, so there is no API key for this app to hold, and
// its /api/model endpoint gives a REAL model list to pick from instead of
// asking the user to type an exact provider-specific model id from memory.

export interface OpencodeModelOption {
  /** "providerID/modelID" — what gets stored and sent back on ask. */
  ref: string;
  /** Human-readable, e.g. "DeepSeek V4 Flash Free (opencode)". */
  label: string;
}

export async function listOpencodeModels(): Promise<OpencodeModelOption[]> {
  const res = await fetch('/api/ai/opencode/models');
  const body = await res.json().catch(() => ({}));
  // 503 carries real setup guidance ("start opencode serve ..."), so surface
  // the server's own message rather than a bare status code.
  if (!res.ok) throw new Error(body?.error || `OpenCode returned ${res.status}`);
  return (body?.models ?? []) as OpencodeModelOption[];
}

export async function chatOpencode(model: string, messages: ChatMessage[]): Promise<string> {
  const res = await fetch('/api/ai/opencode/chat', {
    method: 'POST',
    headers: { 'Content-Type': 'application/json' },
    body: JSON.stringify({ model, messages }),
  });
  const body = await res.json().catch(() => ({}));
  if (!res.ok) throw new Error(body?.error || `OpenCode returned ${res.status}`);
  if (!body?.answer) throw new Error('OpenCode returned no message content');
  return body.answer as string;
}

// ── Retrieval: two legs run in parallel and get Reciprocal-Rank-Fused
// (docs/AI-ASSISTANT-CONCEPT.md §7 steps 2-3), exactly like the doc
// describes — this is real, not a single degraded leg wearing SourceRef's
// clothes:
//   - local FTS: this app's own already-built offline search surface
//     (app/api/offline/search/route.ts). The encrypted SQLite/FTS5 store it
//     reads only exists in Electron's main process — a 404/503 there means
//     "no local index in this session", not an error.
//   - server embedding: app/api/ai/retrieve (lib/ai/retrieval/mail-embeddings.ts) —
//     real JMAP fetch, real Ollama embeddings, real cosine ranking. A 404
//     there means AI_SERVER_BASE_URL isn't configured; anything else is a
//     real failure, logged but not fatal to the question.
// Either leg being absent degrades to the other with no special-casing
// (reciprocalRankFusion handles an empty array leg for free); both absent
// degrades to an unaugmented question, same as before tonight.

import { reciprocalRankFusion } from './retrieval/fusion';
import type { Scored, SourceRef } from './retrieval/types';

export interface AskSource {
  id: string;
  subject: string;
}

export interface AskResult {
  answer: string;
  sources: AskSource[];
  /** True when the question was answered without any retrieved context. */
  unaugmented: boolean;
  /**
   * WHY the answer was unaugmented — the two cases need different words and
   * different user action, and conflating them is actively misleading:
   *  - 'no-index': the local index isn't available at all (not the desktop
   *    app, no keyring, not signed in, or never built). Told to build it.
   *  - 'no-match': the index IS there and answered; this query just matched
   *    nothing. Told to rephrase. Recency questions ("the last mail", "all
   *    mail in July") land here by design: the index ranks by keyword
   *    relevance and has no notion of "latest" or a date range.
   *  - 'augmented': context was found and used.
   */
  retrievalState: 'augmented' | 'no-match' | 'no-index';
  /** True the moment this call consumed a previously-unassigned licensed
   *  seat on the `server` class (lib/ai/entitlement.ts). Always false for
   *  `local`/`public`, which aren't entitlement-gated. */
  seatJustAssigned: boolean;
}

interface OfflineSearchHit {
  id: string;
  jmapAccountId: string;
  title: string;
  snippet?: string;
}

interface OfflineSearchResponse {
  ok: true;
  hits: OfflineSearchHit[];
}

interface ServerRetrieveHit {
  ref: SourceRef;
  title: string;
  snippet: string;
}

interface ServerRetrieveResponse {
  ok: true;
  hits: ServerRetrieveHit[];
}

interface RetrievedContext {
  contextBlock: string;
  hits: Array<{ id: string; title: string }>;
}

/** Set by the most recent retrieveContext() call so askMail can report WHY an
 *  answer was unaugmented. Module-scoped rather than threaded through the
 *  return type because retrieveContext returns null precisely in the case we
 *  need to describe, and a null can't carry a reason. Single-threaded UI, one
 *  question at a time - no interleaving to worry about. */
let lastLocalIndexReachable = false;

async function fetchLocalLeg(question: string, slot?: number): Promise<{ scored: Scored<SourceRef>[]; text: Map<string, { title: string; snippet: string }>; indexReachable: boolean }> {
  const empty = { scored: [] as Scored<SourceRef>[], text: new Map<string, { title: string; snippet: string }>(), indexReachable: false };
  try {
    // `slot` is load-bearing, not optional decoration: the INDEXER writes under
    // the active account's cookie slot (lib/mail-index-client.ts's catchUpIndex
    // passes it), so a search that omits it resolves to whatever account the
    // multi-slot resolver finds FIRST and can read a different - usually empty -
    // account's index. Single-account installs never noticed; a real
    // multi-account/shared-mailbox setup reads the wrong store every time.
    const slotQuery = typeof slot === 'number' ? `&slot=${slot}` : '';
    const res = await fetch(`/api/offline/search?q=${encodeURIComponent(question)}&limit=6${slotQuery}`);
    if (!res.ok) return empty; // 404/503/401 — no usable index this session, not an error
    const body = (await res.json()) as OfflineSearchResponse;
    if (!body.ok) return empty;
    const text = new Map(body.hits.map((h) => [h.id, { title: h.title, snippet: h.snippet ?? '' }]));
    const scored = body.hits.map((h, i) => ({
      ref: { product: 'mail' as const, accountId: h.jmapAccountId, collectionId: '', itemId: h.id, chunkIx: 0 },
      score: 1 / (i + 1), // rank position is all reciprocalRankFusion reads
    }));
    // Reachable even with zero hits: a 200 means the index answered. That
    // distinction is the whole point - "the index isn't there" and "the index
    // is there and this query matched nothing" are different facts the user
    // deserves to be told apart (see AskResult.retrievalState).
    return { scored, text, indexReachable: true };
  } catch {
    return empty;
  }
}

async function fetchServerLeg(question: string): Promise<{ scored: Scored<SourceRef>[]; text: Map<string, { title: string; snippet: string }> }> {
  const empty = { scored: [] as Scored<SourceRef>[], text: new Map<string, { title: string; snippet: string }>() };
  try {
    const res = await fetch('/api/ai/retrieve', {
      method: 'POST',
      headers: { 'Content-Type': 'application/json' },
      body: JSON.stringify({ query: question, limit: 6 }),
    });
    if (!res.ok) return empty; // 404 (server class not configured) or any other failure — degrade, don't fail the question
    const body = (await res.json()) as ServerRetrieveResponse;
    if (!body.ok) return empty;
    const text = new Map(body.hits.map((h) => [h.ref.itemId, { title: h.title, snippet: h.snippet }]));
    const scored = body.hits.map((h, i) => ({ ref: h.ref, score: 1 / (i + 1) }));
    return { scored, text };
  } catch {
    return empty;
  }
}

async function retrieveContext(question: string, slot?: number): Promise<RetrievedContext | null> {
  const [local, server] = await Promise.all([fetchLocalLeg(question, slot), fetchServerLeg(question)]);
  lastLocalIndexReachable = local.indexReachable;
  const fused = reciprocalRankFusion([local.scored, server.scored], 6);
  if (fused.length === 0) return null;

  const combinedText = new Map([...server.text, ...local.text]); // local wins on overlap: it's the more precise leg (BM25 on exact terms)
  const withText = fused
    .map((f) => ({ ref: f.ref, info: combinedText.get(f.ref.itemId) }))
    .filter((f): f is { ref: SourceRef; info: { title: string; snippet: string } } => !!f.info);

  if (withText.length === 0) return null;

  return {
    contextBlock: withText.map((h, i) => `[${i + 1}] Subject: ${h.info.title}\n${h.info.snippet}`).join('\n\n'),
    hits: withText.map((h) => ({ id: h.ref.itemId, title: h.info.title })),
  };
}

export function buildPrompt(question: string, contextBlock: string): ChatMessage[] {
  return [
    {
      role: 'system',
      content:
        "You answer questions about the user's email using only the numbered excerpts " +
        'below as context. Cite sources by their number in brackets, e.g. [1]. If the ' +
        "excerpts don't contain the answer, say so plainly rather than guessing.",
    },
    { role: 'user', content: `${contextBlock}\n\nQuestion: ${question}` },
  ];
}

/**
 * One saved BYOK profile, resolved to an actual key — the caller picks which
 * profile answers *this* question (docs decision 2026-08-05: several keys,
 * selected case by case, not one fixed "the" public provider).
 */
export interface ResolvedPublicProfile {
  baseUrl: string;
  model: string;
  apiKey: string;
}

export interface AskConfig {
  provider: 'local' | 'server' | 'public' | 'opencode';
  localBaseUrl: string;
  localModel: string | null;
  serverModel: string | null;
  publicProfile: ResolvedPublicProfile | null;
  opencodeModel?: string | null;
  /** Cookie slot of the account whose local index should be searched. Omitting
   *  it reads whichever account the resolver finds first — see fetchLocalLeg. */
  slot?: number;
}

export async function askMail(question: string, config: AskConfig): Promise<AskResult> {
  if (config.provider === 'local' && !config.localModel) {
    throw new Error('No local model selected');
  }
  if (config.provider === 'server' && !config.serverModel) {
    throw new Error('No server model selected');
  }
  if (config.provider === 'public' && !config.publicProfile) {
    throw new Error('No provider profile selected');
  }
  if (config.provider === 'opencode' && !config.opencodeModel) {
    throw new Error('No OpenCode model selected');
  }

  const retrieved = await retrieveContext(question, config.slot);
  const messages = retrieved
    ? buildPrompt(question, retrieved.contextBlock)
    : [{ role: 'user' as const, content: question }];

  let answer: string;
  let seatJustAssigned = false;
  if (config.provider === 'public') {
    const profile = config.publicProfile as ResolvedPublicProfile;
    answer = await chatPublic(profile.baseUrl, profile.apiKey, profile.model, messages);
  } else if (config.provider === 'server') {
    const result = await chatServer(config.serverModel as string, messages);
    answer = result.answer;
    seatJustAssigned = result.seatJustAssigned;
  } else if (config.provider === 'opencode') {
    answer = await chatOpencode(config.opencodeModel as string, messages);
  } else {
    answer = await chatLocal(config.localBaseUrl, config.localModel as string, messages);
  }

  return {
    answer,
    sources: (retrieved?.hits ?? []).map((h) => ({ id: h.id, subject: h.title })),
    unaugmented: !retrieved,
    retrievalState: retrieved ? 'augmented' : lastLocalIndexReachable ? 'no-match' : 'no-index',
    seatJustAssigned,
  };
}