One-install front door for a small business that keeps its own books on QuickBooks Online. Install this first: it lights up locally at cold start with first-run onboarding, a Recued-native setup profile, and a unified work queue that pulls inbound email, calendar, and follow-ups into one prioritized list, plus mail-attachment invoice intake that parses a received invoice/receipt with the docling document toolkit and stages the extracted fields preview-first. It deepens when you bind QuickBooks and a Google Drive receipts folder: the daily back-office becomes a live AR / overdue / cash-in cockpit, overdue-invoice-chase drafts one reminder per past-due customer and sends the owner-approved batch through the mail-send gate, and the Drive folder-watch lane stages each new document through the same docling parse. Born blocked, honestly: before the QuickBooks and Google Drive binds the local queue and mail-attachment intake still run, while the accounting and Drive lanes show "connect QuickBooks / Google Drive." Capability packs own the execution points and gates - every QuickBooks write (invoice/payment/refund/void) and every Google Drive folder write stays approval-gated in accounting-quickbooks and gdrive, and customer email only ever sends owner-approved. This pack owns the working operating model; the capability packs own the money and the actuation. Shipped pinned to QuickBooks; collapses to the vendor-agnostic smb-finance-starter once Xero lands.
Marketplace
Inspect what a workflow does, what it installs, and what it can access before bringing it into Recued.
927 packs
One-install front door for a small business that keeps its own books on Xero. Install this first: it lights up locally at cold start with first-run onboarding, a Recued-native setup profile, and a unified work queue that pulls inbound email, calendar, and follow-ups into one prioritized list, plus mail-attachment invoice intake that parses a received invoice/receipt with the docling document toolkit and stages the extracted fields preview-first. It deepens when you bind Xero and a Google Drive receipts folder: the daily back-office becomes a live AR / overdue / cash-in cockpit, overdue-invoice-chase drafts one reminder per past-due customer and sends the owner-approved batch through the mail-send gate, and the Drive folder-watch lane stages each new document through the same docling parse. Born blocked, honestly: before the Xero and Google Drive binds the local queue and mail-attachment intake still run, while the accounting and Drive lanes show "connect Xero / Google Drive." Capability packs own the execution points and gates - every Xero write (invoice/payment/refund/void) and every Google Drive folder write stays approval-gated in accounting-quickbooks and gdrive, and customer email only ever sends owner-approved. This pack owns the working operating model; the capability packs own the money and the actuation. Shipped pinned to Xero; collapses to the vendor-agnostic smb-finance-starter once Xero lands.
Vendor-neutral document-intake pack for a small business that keeps its own books. The two recipes parse received invoices/receipts and stage the extracted fields, with no accounting-vendor binding of their own: invoice-intake-organizer parses one received document with the docling toolkit and stages the extracted fields (preview-first); invoice-intake-watch polls a Google Drive folder for new documents and stages each through the same docling parse (download straight into the warehouse CAS as a file_ref, no bytes through op-step values). The accounting reporting workflows (overdue-invoice-chase, accounting-daily-back-office) moved to the accounting capability packs (accounting-quickbooks / accounting-xero) where they resolve their canonical op-steps against the bound vendor — install one of those for the AR / overdue / cash-in side. Bind the Google Drive connection once and assign each recipe a dish.
End-to-end one-person-company starter pack. Install this first when the business runs on HubSpot, Stripe, email, calendar, and reception: it gives the owner first-run onboarding, a Recued-native setup profile, a unified work queue, weekly owner review, customer support reply desk, outbound follow-up loop, daily command center, inbound mail triage, task and commitment extraction, meeting context, CRM pipeline views, CRM maintenance writes, HubSpot engagement alerts, Stripe billing back-office workflows, and the Stripe money capability under the same install story. Capability packs own execution points and gates; this pack owns the working operating model.
Pick this for the daily cockpit: a set of cards — today's schedule, open commitments, stalled projects, recent notes, inbound-mail triage, and meeting prep — built over your own tasks, notes, and contacts. No external CRM, billing, or GitHub account is set up here; add those with sibling packs when you need them. Choosing between the solo desks? Solo Discipline Loop installs these same cards plus an email outbox so replies and sends work end to end; Solo Work Queue shows everything as one merged list instead of separate cards. Start here if you want the daily view and will connect mail later.
One-person-company developer pack for founders and indie hackers whose work arrives as GitHub issues plus mail. It installs the GitHub-to-Codex background workflow and the personal work queue views that make those runs visible alongside ordinary commitments.
Pick this for a self-contained operating system with no external vendor account: everything in Solo Command Center (today's schedule, open commitments, stalled projects, recent notes, meeting prep) plus an email outbox, so inbound mail is captured into tasks and commitments and replies can be sent. Built entirely on Recued's own tasks, notes, and contacts — add a CRM, billing, or GitHub pack later when the business needs vendor depth. If you don't need the mail outbox yet, Solo Command Center installs the same daily cards without it.
Preview-first onboarding pack for a one-person company. It captures the owner's operating profile, shows the setup checklist, and can seed Recued-native notes, projects, tasks, and a weekly review commitment when the owner explicitly switches write_mode to create.
Pick this if customers reach you while you're heads-down on other work: a front desk that takes in requests through a Reception form and your inbox, holds every submission for your review instead of acting on its own, sorts mail into action vs. noise, and turns the requests you accept into commitments or tasks — so nothing becomes another inbox to babysit.
One-person-company revenue pack for a HubSpot-backed business with Stripe billing. It turns the CRM and billing capability packs into a daily operating surface: open pipeline, forecast, top deals, contacts needing cleanup, CRM maintenance writes, account summaries, engagement alerts, invoice recovery exceptions, payment exceptions, and customer billing cleanup.
One-person-company revenue pack for a Salesforce-backed business with Stripe billing. It installs the same daily operating surface as the HubSpot variant over Salesforce and Stripe capability packs: open pipeline, forecast, top opportunities, contact cleanup, CRM maintenance writes, account summaries, engagement alerts, invoice recovery exceptions, payment exceptions, and customer billing cleanup.
Pick this if you want one prioritized list instead of separate cards: a single work queue that automatically pulls inbound email, calendar activity, and contact follow-ups into tasks, commitments, and reminders, so everything needing attention shows up in one place. It is the queue-first way to start the back office — add external CRM, billing, GitHub, and approval lanes with sibling capability packs. Prefer a dashboard of separate daily cards (today, commitments, stalled projects, notes)? Choose Solo Command Center or Solo Discipline Loop instead.
V6 one-install Stripe API front door over 12 bounded feature packs. The visible root grows from 36 to 96 operations while preserving every established operation ID, route, argument, entity, risk, approval, pagination, and cache contract; the only compatibility hardening is lossless parsing for unsafe JSON integers. Dependencies bring the installed family to 610 of 619 official public GA operations; 9 exact deprecated or unsupported request/response-media routes remain fail-closed. New calls pin Stripe-Version 2026-06-24.dahlia, require approval for every mutation, keep reads uncached, and require caller-owned Idempotency-Key values for POST retries. Expand and Stripe-Account are omitted, form writes use bounded closed typed subsets, v2 JSON stays verbatim, and unsafe response integers remain exact strings. Basic credentials come only from the enrolled Stripe connection. Secret-capable direct responses are labelled and never bundled into workflows. Stripe control-plane records are not claimed as user-work Sources.
Read-only weekly review pack for one-person companies. It composes the Solo Work Queue and Recued-native discipline topics into a Monday owner review: open loops, stalled projects, projects without next actions, commitment reliability, warm notes, meeting context, and relationship context. This pack owns the review ritual, not any new execution primitive.
V3 deterministic audio-parse capability pack. By-value connector composition (service_kind=cli, no separate ingredient): the local whisper CLI transcribing a recorded audio file (mp3 / wav / m4a / flac / ogg — whisper auto-decodes via ffmpeg) into plain text the model can read, via the audio.transcribe catalog operation. This is the media-toolkit parse-in flagship — audio -> text mirrors docling's document -> Markdown. Fixed-function and reproducible: transcription happens locally before any model sees the content. Write-tier but approval=never: the transcript is a faithful local file write into an engine-managed throwaway temp dir, ingested into data.file.received (CAS) and surfaced as result.file_ref; the temp dir is removed after capture, so the only durable copy is the content-addressed, Gateway-gated data.file record (output_capture). A downstream ai-* step reads the transcript via { file_ref } so the content never flows through op-step values. The source can be a data.file ref (the ffmpeg media.extract_audio handoff, or a Drive/manual download) or a local path (input_materialize). Long recordings (whisper runs an ML model) can exceed the 90s foreground cap; those ride a detached execution path when it lands. Requires the openai-whisper CLI (pip:openai-whisper) and ffmpeg on PATH (whisper shells out to ffmpeg to decode); the model weights download on first run. Part of the Lane-1 media toolkit alongside ffmpeg (extract audio) and imagemagick (convert images).