Omniscio

Built-in AI helpers (how they're powered — paid cloud vs free BYO key)

The small background AI helpers Omniscio runs for you — session titles, suggestions, reply drafts, the daily digest, email summaries — what powers them on each plan, what the monthly allowance actually is, and what the "helpers are paused" inbox cards mean.

On this page

What it is

Omniscio runs a set of small built-in AI helpers in the background — session titles, AI suggestions, reply drafts, the daily digest, email summaries, and similar utility calls. These are separate from your coding sessions (which run on your own Claude / Codex / other accounts) and from your prepaid jls_sk_ API keys (see api-keys.md). They are cheap utility-model calls that keep the app's small conveniences working.

Where to find it

Two places matter, and neither has a screen of its own. Your own OpenRouter key — the thing that makes the helpers run on your account — is entered in Settings → Accounts & Providers → API Keys, and your plan together with your live usage and next reset date is on Settings → Plan & Usage. The helpers themselves just run in the background.

How it behaves

How they're powered — your key order (since 2026-10-06)

Each helper call tries a short ordered list of who can pay for it, top to bottom, and uses the first one that works. Out of the box the order is:

  1. Your OpenRouter key (an sk-or-… value you pasted in Settings → My API Keys).
  2. Your Claude key (an Anthropic API-key account — a Claude subscription login can't be used here).
  3. The Omniscio AI allowance — the company-paid cloud lane that comes with a paid plan.

DeepSeek is never on this list (since 2026-10-25). The helpers summarize email, calendar events and documents, and Google's rules for its user data forbid sending that to DeepSeek, so a saved DeepSeek key is not used for them and a DeepSeek row or pick saved earlier is ignored.

That holds for every built-in helper, not only the four jobs on this list. Supermail Plain Speak, the inbox-review stand-in, the voice-match judge in the Voiceprint studio's blind test and the classifier that routes incoming bug reports to their owner (on your computer and in the cloud) all run on GPT-5.6 Luna through OpenRouter. The Background AI presets never put a DeepSeek row in the helpers' order, and a saved DeepSeek key is never shown as paying for the helpers. Because the voice-match judge now answers on Luna, your own OpenRouter key can pay for a blind test.

A row you haven't set up is simply skipped, so with no keys of your own a paid plan runs on the allowance exactly as before. If a key is refused or out of credit, the same request moves to the next row and the helper still answers; a request that timed out never moves on, because the first provider may already have charged for it.

You can change the order — including putting the allowance first again — and pick a key per job (Session titles, Suggested replies, Reply drafts, Summaries & briefings) under Settings → My API Keys → "Which key pays for AI extras". A per-job pick goes first for that job only; the order still backs it up. A key that can run the job's usual model runs it; otherwise it runs its standard one (Claude → Haiku). Free plan, no own key → the helpers pause with a "Built-in AI helpers are paused" card; there is no silent fallback onto Omniscio's cloud and no bundled key.

Plain Speak is the one exception today: its pipeline talks to OpenRouter directly, so it uses your OpenRouter key when set and the allowance otherwise.

Signed-out or an unrecognized plan is treated as free (bring your own key) — conservative, so the company cloud is never handed to an unverified plan.

What the limit actually is

The allowance is a money ceiling per user per month, not a request count — PER_USER_MONTHLY_USD_BY_TIER in paid-offering.ts, enforced by the gateway:

Plan Monthly built-in-AI allowance
Free $0
Pro $0
Team $0
Enterprise $0

As of 2026-08-25 that allowance is $0 on every plan (no free company-funded AI budget on any tier. In practice this means the built-in helpers always run on your own key: general in-app AI declines with a 402 and falls back to your OpenRouter/Anthropic key or your prepaid credit. Three safety-net features — the inbound-email security screen, the keyless Ask-Omniscio help and the new-user onboarding sessions — are exempted from the per-user cap in the gateway (up to a small daily allowance per person) and still run on the shared company pool. Together they are a one-time $1 starter allowance per account, and at most $3 per install across every free account that signs in on the same computer; past either, the feature answers "Your free starter credit is used up. Connect your Claude account to keep going." Paid plans are never held to the per-install limit.

Session titles are included on every plan. A free plan gets 250 titles a month (since 2026-11-09): every new session is named by the company lane — the same pinned model a paid plan uses — instead of from the first few words of your message. Past 250 in a calendar month, or once the allowance is switched off, new sessions are named from their first words again. Someone who has brought their own OpenRouter key keeps paying with it and is not counted against the 250.

On Pro, Team or Enterprise with no key of your own, titles run on Omniscio's cloud even though the allowance is $0 — about $0.0002 a title. Titles have their own per-person ceiling of $1 a month (DEFAULT_PAID_INCLUDED_MONTHLY_USD in company-ledger.ts) — thousands of titles — and the per-person daily limit and the company-wide caps still apply. Past the ceiling a title quietly falls back to the session's first words. On the free plan with no key of your own, a session is titled from its first words instead.

How the free company AI is protected from account farming (since 2026-11-04)

Every new account gets a little company-funded AI (the onboarding sessions, the keyless help, the email safety screen), so the gateway — never the app — holds these lines:

  • Each feature only buys its own models. A feature tag can name only the models that feature really uses, and the email screen and session titles must also send a short, plain request. A forged tag cannot spend its allowance on anything else.
  • Small per-person grants. The email screen's lifetime grant is $5 (years of real use).
  • Shared budgets for the free tier. The free tier as a whole, each network (an IPv4 /24 or an IPv6 /56), and each device have monthly budgets, on by default. Voice dictation is held to the same three.
  • New accounts unlock at a steady pace. An account less than a day old gets company AI only after its network admits it into that day's quota of new accounts (10 by default). A real household or office rarely notices; a burst of fake accounts on one network is held back.
  • An optional card requirement. One server switch, off by default, can require a card-backed free trial before free company AI unlocks.

The meter still resets at the first instant of the next UTC month, and two ceilings sit above the whole lane: a monthly cap on this lane and a Cloud Billing budget. Your live usage and reset date are on Settings → Plan & Usage.

Read the lane cap from the code, never from a number typed into a page. The two ceilings do different jobs and are easy to conflate:

  • the lane cap is a hard stop — over it the lane 402s and calls fall back to your own key. Its shipped value is DEFAULT_GLOBAL_MONTHLY_USD in company-ledger.ts.
  • the Cloud Billing budget is alert-only — it emails and never halts spend. Its value is GATEWAY_BUDGET_MONTHLY_USD in deploy.sh.

A concrete figure on this page would go stale the moment either cap moves, which is why the page states the shape and points at the declaration instead.

A missing entry for a known plan means uncapped, not "no allowance" — which is exactly why every plan that has no allowance is listed explicitly at $0 rather than removed. Never delete a plan's entry to express "this plan gets nothing"; that silently makes it unlimited. (developer is the one deliberate omission: its absence is how the developer subsidy stays uncapped.)

A plan name the manifest does not define — a retired starter, a blank, a forged claim — is a different case: it resolves to Free, so it takes Free's $0 allowance and Free's features. It can never read as unknown, and never as uncapped.

The client-side daily safety net (a runaway brake, not your budget)

Separately from the gateway's per-user monthly ceiling above, the shared helper path carries a client-side daily net: once a helper has spent $25 in one local-midnight day under its own label, that helper stops calling and resumes the next day. Session titles, suggestion chips, reply drafts, hub/app/skill descriptions and the other small assists all run through the one shared path, so this is the net under all of them.

Read what it is before treating it as a budget. The constant describes itself as "a safety net, not a tuning knob" and is deliberately high — normal use of these helpers is cents a day, so the net exists to stop a genuine runaway (a supervisor loop re-firing on a stuck session all day), not to meter you. It is counted per label, per cost source, summed across accounts, and it is not the per-user spend ceiling: when a helper routes through the company gateway, the authoritative per-user ceiling is enforced server-side there and answers 402 before the call is forwarded. This is also a different limit from the two "helpers stopped" cards below, which are about the monthly allowance.

The two "helpers stopped" cards are different

One inbox row (dedupKey pooled-ai-monthly-limit), but two honest messages — they must never be swapped, because only one of them involves anything actually being spent:

  • "Built-in AI helpers are paused" (no-plan-allowance) — the free-plan pause above. Nothing was consumed and no allowance ever existed, so this card never says an allowance was "used". It offers both routes out: upgrade, or paste your own key.
  • "Add your own API key" (pooled-ai-monthly-limit) — a real gateway 402. Since 2026-10-08 (ticket f31a8de8) this card no longer claims an allowance was used: with every tier at $0, no allowance for these helpers ever existed, and "you have used your $0" was the same species of lie the two variants exist to prevent. It now says what is true — the helper needed a key of your own — and points at the key box. A keyless caller gets ai-no-api-key-notice ("Add an API key to turn on AI helpers") instead, so the two never contradict each other.

The gateway's own 402 speaks the same way (2026-10-26). When the server refuses a call at the per-person monthly line it answers "This feature is not company-funded on any plan, so it runs on your own API key or your Omniscio credits. Add your own key to keep using AI features." — it names the funding model and the two real ways forward, and promises no upgrade, because no plan funds that feature. (It says this feature, not any company AI: session titles on a paid plan and the safety-net features ARE company-funded.) The desktop recognises that wording as a billing state you can act on — never an app fault — and still recognises the old "monthly … allowance / Upgrade to Pro" sentence, so an app newer than the deployed gateway keeps working.

Both name the helpers that stopped (session titles, suggested replies, Plain Speak summaries, the daily digest, email summaries) and confirm your Claude coding sessions, projects, inbox and automations are unaffected. The button opens Settings → Plan & Usage. The tier behind the copy is read with currentEffectiveCachedTier(), so signed-out and lapsed-paid users both get the free card with its upgrade path — see internal-ai-gateway-routing-contract.md I9.

The card lands at most once a month

The allowance is monthly, so the notice is too — it cannot reach the user more than once per billing period, which takes two separate caps because the producer fires constantly by design (every helper that touches the spent pool 402s and calls it):

  • Within an app run — a once-per-run latch in pooled-ai-limit-notice.ts. Without it each 402 coalesced into the live row, and because bumpDedupCount touches updated_at while the inbox lists ORDER BY updated_at DESC, the same card re-floated to the top all day.
  • Across restarts and dismissals — a 30-day floor on pooled-ai-monthly-limit in the central alert cadence registry. On the generic 24h default a dismissed card came back the next day, every day, for the rest of the month.

Past 30 days it re-raises normally: a monthly cap must not become permanent silence, because the pool resets and the user is entitled to hear about the new month's.

"Your <key> was refused"

When one of your own keys is refused or runs out of credit, one card per key says so and offers Open My API Keys. The helpers keep working on the next key in your order in the meantime. Unlike the outage card below, this one does not fix itself — check that key and its balance. It clears the next time that key works, even after a restart.

A used-up allowance is never reported as an outage ("usually recovers on its own"): the monthly allowance card appears only when nothing else in your order could carry the request — if your own key carried it, nothing is paused and no card is shown.

Adding your OpenRouter key (free plan)

Create a key at openrouter.ai/keys, then paste it into Settings → Accounts & Providers → API Keys → "Your OpenRouter key". It's stored encrypted like every other credential (never synced in plaintext, never sent to the renderer or a CLI export). Once saved, the built-in helpers run on your OpenRouter account — Omniscio maps each helper's model to its OpenRouter equivalent automatically, so you don't pick models. Your key takes priority the moment it's set.

The old "Route internal AI through the cloud (test)" toggle is gone

Through mid-2026 an experimental toggle ("Route internal AI through the cloud (test)") let you flip the helpers between the bundled keys and your cloud account. That toggle has been removed — routing is now decided by your plan, not a manual switch, so there's nothing to toggle. Paid plans get the cloud automatically; free plans bring a key. (The old setting name routeInternalAiThroughGateway still exists internally as an inert default and is no longer user-facing.)

If you haven't added an API key yet (new installs)

The first time a helper can't run because no usable Anthropic API key is saved, you get one inbox card that says exactly that — "Add an API key to turn on AI helpers" — with an Add API key button that opens Settings → Accounts and scrolls straight to the Anthropic key box. It does not blame a provider outage, because nothing is down: the key was simply never added. A claude login account alone is not enough here — the Messages API these helpers use doesn't accept OAuth tokens, so they need a key even when your coding sessions are signed in and running fine.

You get one card, not a stream: it's raised at most once per app run and coalesced under a single dedup key, so dismissing it sticks. Everything else keeps working while it's up, and it clears itself the moment the helpers succeed after you paste your key.

Before this shipped, the Messages-API helpers refused silently on a keyless install — no card, no explanation, nothing to click — and the only card you could eventually get came from an unrelated code path and blamed a temporary outage.

If the main AI provider has a bad stretch

The built-in helpers run on a cheap, shared AI provider. If that provider has a rough patch (an outage, or its account running low), you might briefly see an "AI features are using a backup provider" notice in your inbox. Nothing is broken — the helpers automatically shift to a backup so titles and suggestions keep working; they may just be a little slower or, now and then, a touch less polished. The notice clears itself once the main provider recovers, and there's nothing you need to do. (And if the AI titler can't reach any provider at all, a new session simply gets a short title made from your first message instead of a smart one.)

What each plan includes, and how to change it, is on the plan and billing page. The coding accounts these background helpers are separate from are on the API keys page, and the daily digest — one of the helpers named here — has its own credential and source rules on the Daily Digest page.

Last verified 2026-11-04