f121678e2af17d20accffa0d5309c2f5f70ebfb1
5
Commits
| Author | SHA1 | Message | Date | |
|---|---|---|---|---|
 Bernd RodlerandClaude Sonnet 5
|
f121678e2a |
feat(ai): Paperclip-style env-var provider presets + zero-config local default
Two product decisions from tonight: 1. Public AI providers can now be published by an admin as named presets (lib/ai/types.ts's PublicAiPreset: name/baseUrl/model/apiKeyEnvVar). The admin names an env var, never a secret value - the actual key is whatever ops has set in the server's real environment, same custody model as the existing AI_SERVER_BASE_URL var. A new server route (app/api/ai/public/chat) resolves it and makes the call itself, which also sidesteps the CORS/wrong-base-URL failure class chatPublic hit earlier tonight. Users pick a preset from a dropdown in Settings - Answer with - no key field at all; personal BYOK (paste your own key) stays available as a secondary "Add your own key" option, not removed. Admin UI: new "Public - org-managed presets" card in the AI policy tab. 2. AI now defaults ON instead of requiring setup (lib/ai/auto-provision.ts): on first load, if no provider is chosen yet, probe OpenCode (this app auto-spawns `opencode serve` itself, so it's the one local option with zero external install step) then Ollama via the existing auto-discovery, and adopt whichever answers. Never overrides an explicit choice - only fires while provider is still null. Wired into both AI entry points (the Ask button and the Settings pane) so it resolves before either renders its "not configured" state. Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com> |
||
|
Bernd Rodler
|
98dcd3b1e9 |
feat(ai): OpenCode provider class; fix retrieval reading the wrong account's index
Three things, all from running the real thing rather than trusting a status code.
1. OpenCode as a 4th AI class (lib/ai/opencode.ts + app/api/ai/opencode/*).
A locally-running `opencode serve` — the same runtime Paperclip drives as
an adapter. Its appeal over a BYOK profile is precisely what was broken
before: opencode owns provider auth itself, so there is NO api key for
this app to hold, and it reports a REAL model list (25 on this machine)
instead of asking the user to type an exact provider-specific model id
from memory. Typing "Sonnet 5" into a free-text box and getting a bare
"Provider returned 401" is the failure this removes.
IMPORTANT trap, documented in the module header and pinned by a test:
opencode is NOT OpenAI-compatible. `/v1/models` and `/v1/chat/completions`
both answer 200 — because a web-UI catch-all serves index.html for ANY
unknown path. I built the first version against that assumed compatibility
on the strength of two 200s and had to throw it away once I read a body.
Every probe now validates the parsed shape and content-type, never the
status alone. The real API is GET /api/model + POST /session +
POST /session/{id}/message, and the reply's `reasoning` parts are stripped
so a model's private chain of thought can never surface as the answer.
Proxied through our own backend (like the `server` class) because the
desktop renderer's origin is a random port that changes every launch;
same-origin sidesteps opencode's CORS allowlist entirely. Loopback-only by
construction: a non-loopback OPENCODE_BASE_URL is refused, since "local,
no keys, nothing leaves the device" is the whole point of this class.
2. Retrieval read the WRONG ACCOUNT'S index. The indexer writes under the
active account's cookie slot (catchUpIndex passes it) but fetchLocalLeg
omitted `?slot=`, so search resolved to whichever account the multi-slot
resolver found first. Single-account installs never noticed; a real
multi-account/shared-mailbox setup reads an empty store every time. Both
call sites now pass the active slot.
3. "No local mail index available in this session" was shown even when the
index existed and simply matched nothing — actively misleading, and it
masked the missing-SESSION_SECRET bug for hours. AskResult now carries
retrievalState ('augmented' | 'no-match' | 'no-index') and the two cases
get different words: build the index, versus rephrase (with the honest
caveat that keyword search answers content questions better than recency
ones like "the last mail").
Verified live against real opencode 1.18.14: discovery found 25 models and a
real prompt round-tripped the exact expected answer through the real helper
code, not curl. Gate: tsc clean, eslint clean, 2512/2512 unit tests (10 new,
incl. one that fails if the HTML catch-all is ever accepted as an API), build clean.
|
||
|
Bernd Rodler
|
30e5059b94 |
feat(admin): build the AI Policy console (§6) — approved, spec now implemented
New admin tab "AI" (app/(main)/admin/_tabs/ai-policy.tsx): provider-class toggles, server model allow-list, BYOK provider allow-list, seats/usage (front-end for the already-real lib/ai/entitlement.ts), retrieval on/off, consent text + version bump. Real backend, not cosmetic: AiConsoleConfig persisted via config-manager (lib/ai/types.ts, ai-policy.json in the CONFIG dir). New GET/PUT /api/admin/ai/policy. Enforcement wired at every real chokepoint, not just the picker: /api/ai/server/chat checks classesEnabled.server and the model allow-list, /api/ai/retrieve checks retrievalEnabled, /api/ai/server/models filters by allow-list. GET /api/ai/policy folds classesEnabled into the classes list clients see. Resolved the spec's 3 open questions as recommended: BYOK allow-list stays client-side/advisory (wired into ai-assistant-settings.tsx's addProfile), tier picker stays cosmetic, master aiAssistantEnabled toggle stays in the existing Policy tab (this tab links to it instead of duplicating it). Defaults preserve today's behavior exactly (classesEnabled/allowlists all start empty/null) — turning this on changes nothing until an admin touches it. |
||
|
Bernd Rodler
|
dda7adf565 |
feat(ai): multi-key BYOK, real server class, real entitlement enforcement
Three pieces built together tonight since they're naturally linked (the
server-class proxy is the real entitlement enforcement chokepoint):
1. Multi-key BYOK (public class): several named provider profiles
(name/baseUrl/model), each with its own key in lib/ai/key-store.ts
(keyed by profile id, not a single fixed 'public' slot). The "Try it"
pane lets you pick which saved profile answers each question - not one
fixed default.
2. `server` class, real: app/api/ai/server/{models,chat} proxy through
this app's own backend to AI_SERVER_BASE_URL - same-origin from the
browser, no CORS/OLLAMA_ORIGINS story at all, standing in tonight for
VNC's EU/CH-hosted infra with the real Ollama on this Mac (swapping to
the real instance tomorrow is a config change).
3. Real entitlement enforcement (lib/ai/entitlement.ts), scoped to `server`
only (not local/public, per the 2026-08-05 decisions): checkAndAssignSeat()
re-validates on every /api/ai/server/chat call - first use auto-assigns a
seat if any remain, further calls from an unlicensed user get a 402 with
a specific reason. recordUsage() appends to an append-only metering
ledger (timestamp/user/model/tokens/latency) that IS the billing record.
Admin data endpoints at /api/admin/ai/entitlement (seat total, revoke) -
the visual admin console is a separate, not-yet-built task.
Two real bugs found and fixed during verification, not just claimed fixed:
- /api/ai/policy never actually added 'server' to entitlement.classes even
when AI_SERVER_BASE_URL was set (only the type comment was updated) - the
Server radio option silently never appeared until this was caught live.
- The new routes used readStalwartAuthContext(0) (hardcoded slot, SSO/reauth-
specific) instead of getStalwartCredentials() (the general multi-slot
session resolver every other authenticated route uses) - reachable but
wrong, and would have hidden a real auth gap behind "works on my slot".
Verified end-to-end for real: built + ran the actual server, logged in via
the real (non-demo) auth flow, selected Server, listed the real Ollama
models through the proxy, asked "Reply with exactly the words: SERVER CLASS
WORKS" and got back exactly that - plus confirmed on disk (not just in the
UI) that data/admin-state/ai-entitlement.json recorded the seat assignment
and ai-metering.jsonl recorded real prompt/completion token counts and
latency from the actual model call. Rejection-path logic (seat limit
reached, zero seats configured, revocation) covered by 5 new unit tests
rather than a second live round trip. Full suite: typecheck clean, lint
clean, translations 48/48, production build succeeds.
|
||
|
Bernd Rodler
|
2a35019b21 |
feat(ai): P0 client scaffolding — capability flags, settings pane, policy fetch
Per docs/AI-ASSISTANT-CONCEPT.md §12, P0 is deliberately generation-free: prove platform gating and the policy round trip before any model exists behind it. No provider is called anywhere in this change. - lib/platform-capabilities.ts: supportsLocalLlm/localLlmNeedsCorsSetup, mirroring the same-named module in vncmail-native so the capability contract (§3, §11) reads identically on both clients. Web+Electron only here — mobile is a separate codebase. - lib/ai/types.ts: AiPolicy/AiEntitlement schema, locked in now per decision #4 (entitlement from day one — cheap now, a live-tenant migration later). - app/api/ai/policy/route.ts: GET, unauthenticated (users read this, like /api/admin/policy). Composes the real FeatureGates.aiAssistantEnabled toggle with a hardcoded unlicensed entitlement — there's no seats/billing backend yet (P2), so nothing here can honestly claim otherwise. - components/settings/ai-assistant-settings.tsx: fetches that policy, shows a real (not fake) locked/unlicensed state. No model config UI yet — there is nothing real to configure until P1/P2/P5 land. - New admin FeatureGates.aiAssistantEnabled (default false, like pluginsEnabled): the tab is entirely hidden until an admin opts in, so no existing install suddenly sees a tab that does nothing. Verified: typecheck clean, lint clean, translations test passes (48/48), full production build succeeds with /api/ai/policy compiled in. |