alfred_
Also known as: alfred_ AI, get-alfred.ai, alfred AI executive assistant
AI executive assistant that triages the inbox overnight and scores messages on sender persistence as well as urgency, then drafts replies in the user's voice that never send without approval.
alfred_ is an AI executive assistant for email, calendar and tasks, aimed at founders, consultants and independent professionals who bill for their time. It is built around working unprompted. It triages the inbox overnight and has a Daily Brief ready by morning.
The triage is a scoring pipeline, not a simple sort. Every message is rated on urgency language, action required phrasing, sender importance including VIP status and executive titles, the age of the message, and how many follow ups that sender has already sent. Revenue critical and chased threads are escalated, and newsletters and notifications are archived automatically. Follow up state is tracked across threads and flagged with rising urgency at the first, second and third approach, so a politely repeated request surfaces rather than sinking.
For anything needing a reply, alfred_ writes a full draft from the thread context and the user's own writing style. Action items are pulled out of messages automatically and become tasks linked back to the source email, with stale and overdue work escalated into the brief. The calendar side calculates meeting load, detects conflicts and scores genuinely available focus time, discounting short gaps between meetings as unproductive. Work is organized in a Kanban board with notes alongside.
For subscribers the assistant maintains a searchable index of recent mail, storing cleaned message text with embeddings so it can answer questions by meaning and cite the message it drew on. The index is encrypted, scoped to the account, never used for training, and deleted with the account.
Oversight is a designed feature. Drafts never send themselves. They wait for review and go with one tap. Creating calendar events or sending scheduling mail from chat is confirmed first. Routine calendar responses, such as accepting an invitation that fits, are handled automatically by default, and a setting switches that to suggest only. Mail that alfred_ sends unattended comes only from automations the user builds (forwarding rules, scheduled sends and workflows), which can be paused or deleted at any time.
It connects to Gmail, Outlook, Google Calendar and Outlook Calendar over OAuth, with several accounts of each supported and unified into one queue, plus Drive and OneDrive file access and SMS as both an alert and an instruction channel. There is no Slack or CRM connector, no meeting transcription, and no public API, SDK or developer surface. Inference runs on the Anthropic, OpenAI and Google APIs, and the customer does not choose among them.
Data is encrypted at rest and in transit on Supabase hosted PostgreSQL, isolated per user at the database layer, and never used to train models. The vendor holds no security certification of its own and publishes no data residency option. Pricing is a single flat tier at $29.99 per month or $299.99 per year, with a team plan at $34.99 per seat and a two seat minimum, sold self serve on a 7 day trial.
Vendor details
Canonical URL
https://get-alfred.ai
Category
Enterprise operations agent
Subcategory
AI executive assistant for inbox and calendar
Funding status
Private, with no funding detail found. Positioned explicitly against the cost of a human executive assistant, at $29.99 per month rather than a $60,000 plus salary.
Company status
independent
Use cases & customers
Primary use cases
Target customers
Deployment options
Integrations
alfred_ connects to both Gmail and Outlook and unifies them into a single triage queue, which is unusual in a category where most tools pick one side. There are no native Slack or CRM integrations yet, so a Slack first workflow still needs a separate tool. SMS is supported as a channel for instructing the assistant.
In practice
The inbox is triaged overnight and a morning brief presents only what needs judgment, with newsletters already archived and chased threads escalated.
A client sends a third polite request for an update and it is surfaced because the assistant scores sender persistence, not just recency.
A reply is drafted in the user's own voice and waits for one tap of approval instead of sending itself.
Sources & related URLs
Related / legacy domains
Agentic Index coverage score
5.5 / 14 capabilities · 39%
| Integrations & Tool Calling | Partial |
|---|---|
|
The connectors are the mail and calendar accounts the product runs on. alfred_ integrates with Gmail, Outlook (Microsoft 365), Google Calendar and Outlook Calendar, and a user can connect one or several accounts of each type. It signs in with OAuth 2.0, with scoped permissions to read and write email and calendar, read contacts only, and access cloud files. Within those accounts the agent writes as well as reads. It archives mail, sends approved drafts, creates calendar events and accepts invites, all as authenticated writes into systems the vendor does not own. Several Gmail and Outlook accounts can be unified in one queue, where most rivals pick one side. Google Drive and OneDrive file access comes in when file features are used, alongside a Zoom connection and SMS as a two way channel. There is no native Slack connector, no CRM integration and no meeting transcription. Beyond the four mail and calendar endpoints there is no task tool, ticketing, storage or messaging connector, and there is no integration marketplace. A Slack first or CRM centered workflow needs a second product. Sourceget-alfred.ai/help Getting Started integrations FAQ, get-alfred.ai/security authentication sectionread 2026-09-13 |
|
| Workflow Orchestration | Full |
|
The work runs unprompted as a pipeline of numbered stages. Emails arrive and are read and classified by urgency, sender importance and action required. Noise is archived automatically. Replies are drafted from the user's communication style and full thread context, and everything needing judgment surfaces in the Daily Brief. Task extraction runs alongside, scanning every email for action items such as commitments, deadlines and requests, and tasks are created and linked back to the source message automatically. Every message is scored on urgency keywords, action required language, sender importance including VIP status and executive titles, the follow up count from the same sender, and email age. Revenue critical and chased threads are escalated. Follow up state is tracked across threads with escalating urgency (first, second, third plus), so the pipeline carries state between runs instead of deciding each message in isolation. Users can also create automations that send on their own, such as forwarding rules, scheduled sends and workflows set to run automatically. These run unattended, persist beyond a session and are managed in Settings, where they can be reviewed, paused or deleted. Calendar analysis runs as a parallel branch, covering meeting load, conflict detection and focus time scoring that discounts short gaps as context switching. There is no visual workflow builder, no branching or conditional composition and no multi agent delegation. Sourceget-alfred.ai/help Email & Triage, Daily Brief & Tasks and Calendar sectionsread 2026-09-13 |
|
| Knowledge Grounding & RAG | Full |
|
For subscribers, alfred_ keeps a persistent, searchable index of recent email so it can answer questions and cite the right message. It holds cleaned message text, with quotes and signatures stripped, plus embeddings that let it search by meaning, and it is maintained over time, not context assembled for each run. Answers point the user to the message they drew on, so retrieval traces to a source message instead of being summarized away. The index is encrypted, isolated to the user's account and never used for training, and it is deleted with the account or on request. Beyond the index, drafts are written from the full thread context, and the assistant synthesizes information from the user's email, calendar and tasks. Cloud files in Google Drive or OneDrive are reachable when file features are used. The index covers recent email, not the full mailbox, and only subscribers get it. Sourceget-alfred.ai/security search index and FAQ, get-alfred.ai/help Email & Triageread 2026-09-13 |
|
| Human Oversight & Guardrails | Full |
|
Every draft waits for the user's approval before it goes out, and a draft never sends itself. This gate is on by default and has no exceptions. Drafts appear ready to review, edit and send with one tap, and they are only sent when the user approves them. Calendar works differently. Creating calendar events or sending scheduling emails from chat is confirmed with the user first. By default, though, alfred_ handles routine calendar responses on its own, such as accepting an invite that fits, and a setting switches this to suggest only mode at any time. The only mail alfred_ sends on its own comes from automations the user sets up, such as a forwarding rule or a scheduled send, which the user can review, pause or delete. Wrongly extracted tasks can be dismissed with one click. There is no admin policy engine, approval queue or per action risk tiering, and the product, built for single seats and small teams, has no administrator role. Sourceget-alfred.ai/help Email & Triage and Calendar sections, get-alfred.ai/security FAQread 2026-09-13 |
|
| Security, Identity & Governance | Not documented |
|
Each compliance claim belongs to a supplier, and alfred_ holds no attestation of its own. Data sits on Supabase hosted PostgreSQL in SOC 2 compliant data centers, and that attestation is Supabase's. Payments are processed by Stripe, which holds PCI DSS Level 1. alfred_ calls its own posture GDPR friendly and aligned with GDPR principles, which is not a certification. No SSO, SAML, SCIM, roles, permission model or customer visible audit log is named. The team tier offers simple seat management, which covers billing administration, not access. Access control at alfred_ means user level isolation at the database layer and session expiry. Internal staff access to production data is strictly limited, logged and requires multi factor authentication. alfred_ runs these controls itself, and customers do not operate them. Sign in to Gmail and Outlook uses OAuth 2.0, and data is encrypted at rest and in transit. Sourceget-alfred.ai/security, get-alfred.ai/help Privacy & Security, get-alfred.ai/pricingread 2026-09-13 |
|
| Observability & Auditability | Partial |
|
Each morning alfred_ prepares a Daily Brief with the emails that need judgment, draft replies ready to send, tasks due today, the calendar and patterns it has noticed. The brief also covers what alfred_ handled overnight. A user can see that newsletters were archived, which threads were escalated and which tasks were extracted, and each extracted task links back to the source email, so an action can be traced to what started it. Follow up state shows as first, second and third plus, and task aging as stale and critical, so the agent's ongoing judgments are visible. alfred_ publishes the five scoring factors but shows no breakdown for each message and no explanation of why a message scored as it did. It describes no decision log or reasoning trace, no run history beyond the current morning's brief, no searchable record of past agent actions, no export and no retention statement. The only logging it describes covers its own staff's access to production data. Sourceget-alfred.ai/help Daily Brief & Tasks and Email & Triage sectionsread 2026-09-13 |
|
| Memory & State Persistence | Not documented |
|
Drafts follow the user's communication style, and alfred_ says the system learns stakeholder hierarchy from email patterns and gets more accurate over the first week. That learning is absorbed into the model, and no one can point at, inspect or edit it. Follow up counts are the product's own tracking state, since alfred_ monitors every thread for response status as part of the scoring pipeline. Chat conversations with alfred_ are stored so the user can pick up where they left off. The record is persistent and per user, but it is a log of what was said. Nobody curates it, and the assistant does not consult it as authored guidance. alfred_ stores OAuth tokens, tasks, notes, preferences, chat history and the search index, and preferences are settings the product applies as rules. It names no store, written in natural language, that the assistant reads as context to decide. There is no scope beyond the account, no stated lifetime, no way to read or write such a store and no way to purge memory apart from deleting the account. Sourceget-alfred.ai/security data privacy and AI processing sections, get-alfred.ai/helpread 2026-09-13 |
|
| Deployment & Data Residency | Not documented |
|
The product runs as cloud SaaS, and alfred_ names no country, region or EU option for where data lives. The production database is hosted on Supabase with enterprise grade PostgreSQL infrastructure, encrypted storage volumes and daily backups. Data sits in SOC 2 compliant data centers, with no country or region given. Customers cannot learn which country their mail is indexed in, and there is nothing to choose. There is no self hosting, private or dedicated tenancy, bring your own cloud or region selection, and no enterprise tier that could sell one. The plans are a single individual seat and a team tier with a two seat minimum. The product syncs a customer's Gmail or Outlook mailbox into alfred_'s own Supabase estate and builds a searchable embeddings index there, and inference goes to the Anthropic, OpenAI and Google APIs. Row level security, per account isolation, AES-256 and TLS 1.3 protect the data but do not change where it lives. Sourceget-alfred.ai/security infrastructure and FAQ, get-alfred.ai/pricingread 2026-09-13 |
|
| Prebuilt Agents, Templates & Packs | Not documented |
|
One assistant is the whole product. Triage, drafting, task extraction, calendar analysis, the Daily Brief, Kanban and Notes are features of that single product, not separate units a customer chooses among. The For Founders, For Real Estate, For Lawyers, For Educators & Coaches and For Accountants pages sit under Built for in the Solutions menu, beside a Compare menu. They are landing pages for those audiences, not configured agents a customer adopts. alfred_ does not say that a lawyer's assistant behaves differently from an accountant's, and none is offered to install. There is no gallery, template library, starter pack, agent directory or marketplace. Automations such as forwarding rules and scheduled sends are the one thing a customer can build, and each is written from scratch. An individual seat costs $29.99 and a team seat $34.99 with a two seat minimum. The team tier includes everything in alfred_ plus billing administration. Sourceget-alfred.ai/pricing plan comparison, get-alfred.ai/help navigation and guides indexread 2026-09-13 |
|
| Triggers & Channel Coverage | Full |
|
Work starts in three ways. New email starts it. alfred_ reads and classifies every new message as it arrives, and archival and drafting follow automatically with no person present. A schedule starts it too, and the overnight run is the product's core promise. Once a user connects a Gmail or Outlook account, the inbox is triaged, draft replies are queued and tasks are extracted by the next morning. An optional 7am SMS delivers the day's calendar and tasks. Changes in state also start work. alfred_ monitors every thread for response status and flags follow ups with escalating urgency. Task aging is watched the same way. Tasks older than seven days are flagged stale, and tasks overdue by three or more days are marked critical and escalated. These fire when something fails to happen over time, which no inbound message or clock tick would produce alone. Customers can add their own triggers, since automations include scheduled sends and workflows set to run automatically. Users reach the assistant through the web app, the Daily Brief and SMS, which works as both an alert channel and an instruction channel. There is no webhook, inbound API or external work queue, and the product has no developer surface. Sourceget-alfred.ai/help Getting Started, Email & Triage and Daily Brief & Tasks sectionsread 2026-09-13 |
|
| Model Flexibility & Routing | Partial |
|
AI processing runs on the Anthropic (Claude), OpenAI and Google (Gemini) APIs under commercial API terms, and customer content is sent to them for inference only. alfred_ lists these three frontier providers as the model providers its own processing runs on, and the customer chooses none of them. Routing between them happens on alfred_'s side. It names no model picker, per request selection, admin entitlement or bring your own key. Sourceget-alfred.ai/security AI processing sectionread 2026-09-13 |
|
| APIs, SDKs & MCP Extensibility | Not documented |
|
There is no public API, SDK, developer portal, MCP server, webhook or CLI, and no developer section or subdomain. Every connection is one alfred_ makes. It calls Gmail, Outlook, Google Calendar, Drive and OneDrive through OAuth, and the Anthropic, OpenAI and Google model APIs for inference. Nothing lets an outside system call alfred_. SMS is an instruction channel for a person, billing runs through a Stripe portal, and alfred_ runs PostHog analytics on its own product. Customers cannot extend the product. Automations such as forwarding rules, scheduled sends and workflows set to run automatically are configured inside the product from fixed options, and no code, callback or custom tool can be added. There are no native Slack or CRM integrations, and alfred_ is a closed product, not a platform. Sourceget-alfred.ai/help full navigation and guides index, get-alfred.ai/securityread 2026-09-13 |
|
| Testing, Debugging & Optimization | Not documented |
|
For checking a change, alfred_ publishes no test harness, scored cases, dry run, preview mode, regression check, A/B comparison or quality metric. The triage engine archives mail on its own, using a five factor score that the customer cannot inspect for each message, tune, or test against past threads before it runs. A misjudged VIP weighting can silently bury a client thread, and alfred_ describes no feedback loop beyond noticing afterwards. Automations that customers create, such as forwarding rules and scheduled sends, are written in Settings, and the first time they run is against live mail. The approval gate on drafts checks each message, not a change to how the agent behaves. Dismissing a wrongly extracted task with one click corrects that one item. The ROI Calculator on the marketing site computes claimed hours saved from the buyer's own inputs before purchase and tests nothing about the agent. alfred_ publishes no accuracy figure for triage classification, no benchmark and no draft acceptance rate. Sourceget-alfred.ai/help Email & Triage and Daily Brief & Tasks, get-alfred.ai/pricingread 2026-09-13 |
|
| Browser & Computer Use | Not documented |
|
The agent drives no hosted or local browser session and has no desktop control and no remote or local computer control. alfred_ lists every external interface the product touches, and none of them is a browser or a computer. Everything runs through APIs. alfred_ reaches Gmail, Outlook, Google Calendar, Outlook Calendar, Drive and OneDrive through OAuth 2.0, with scoped permissions to read and write email and calendar, read contacts only, and access cloud files. It reaches Anthropic, OpenAI and Google through their APIs. Every system it touches has a programmatic interface, so nothing renders a page and clicks it. alfred_ says it reaches nothing outside the accounts a user connects, with no browsing history, no location and no data brokers. Cloud file access to Drive and OneDrive is an API read, not an agent operating a file manager. Speech recognition runs on the user's own device and only the text reaches alfred_, so voice is an input channel. Sourceget-alfred.ai/security authentication, data privacy and AI processing sectionsread 2026-09-13 |
|
The Agentic Index coverage score grades every vendor Full, Partial or Not documented against the same 14 buyer facing capabilities, from public evidence only. Each capability links to how all vendors in the index score on it. How this evidence is graded
Product updates
Version 1.1.21 of the alfred_ iPhone app brings the web Home to the phone, with routines shown as cards, swipeable cards for what needs the user and an Ask alfred_ bar. Two new Home Screen widgets come with it. One counts down to the next meeting with its attendees and a Join button and flags any reply owed to someone on the call, and the other shows the latest result of a chosen routine.
Bears on: Agent capability
View sourcePricing
$29.99 per month, or $299.99 per year. The team tier is $34.99 per seat per month, with a two seat minimum.
Flat monthly subscription with AI included, billed per seat on the team tier.
Included quota
AI email triage, draft replies, task extraction, calendar intelligence and daily briefings are all included in the single price with no metered component.
What is public
Both tiers, both billing cadences, the seat minimum and the trial length are published as exact figures on a self serve pricing page with no sales gate. Some older pages on the same site still carry a different price and trial length.
Cost watchouts
The real cost is coverage rather than metering. With no native Slack or CRM integrations and no meeting transcription, work spanning those surfaces needs a second tool. It is a personal inbox product, so a shared support or sales inbox is a different purchase. The team tier carries a two seat minimum, so the smallest team purchase is $69.98 per month rather than $34.99. And because every draft requires approval, throughput is capped by how quickly the user reviews, and the assistant cannot run further ahead than its approver.
Variable cost rationale
Nothing is metered. The price is a flat $29.99 a month or $299.99 a year with the AI included, and there is no per resolution charge, no AI add on, no consumption unit and no token pass through, so a customer knows the annual cost on day one. The only way the bill grows is adding seats on the team tier, at $34.99 per seat per month with a two seat minimum. No quota, message cap or fair use limit is published, so whether a volume ceiling applies at the flat rate is unknown.
Additional watchouts
The gaps cost more than the subscription. No Slack, no CRM and no meeting transcription means another tool alongside for most teams. The team tier has a two seat minimum, and the approval gate limits throughput, since the assistant cannot get further ahead than the person approving it.
Overage / add-ons
None published. There are no per resolution charges and no AI add on fees, and no overage, quota or fair use ceiling is published.
Sales call required
No, self serve available
Free / trial
7 day free trial on every plan, with a card required, and no free plan
Lowest paid plan
Starter at $25 per month, billed annually
Commercial notes
The AI is included in a flat price with no per resolution charge, no AI add on, no agent consumption unit and no token pass through.
Key ambiguities
The price is $29.99 per month or $299.99 per year, and the help center's Account & Billing section reads "subscriptions are billed monthly at $29.99." Older pages on the same site are inconsistent, and $24.99 persists on /alfred-ai, /ai-executive-assistant and at least two blog posts. Trial length varies the same way. /pricing says "7 days free," one blog post offers a 30 day free trial with no credit card, and the help center says only to "check the pricing page for current trial length." The trial on the pricing page is 7 days. Still open is whether any volume ceiling applies at the flat rate. No quota, message cap or fair use limit is published, and the searchable mail index is described as covering recent email without saying how much.
Cancellation / refund
Cancel anytime with no fees.
Missing data
Whether any volume ceiling applies at the flat rate.
Related vendors
- 11th Estate — Agentic platform whose AI engine scans markets, matches an…
- Abacum — FP&A platform for growing companies whose AI Modeler builds, edits…
- Adopt AI — AI native accounting and bookkeeping firm whose agents run…
- Aera Technology — Decision intelligence platform where an always on agent executes…
- Akro AI — On premise operational intelligence that automates document heavy…
- Alloy.ai — Commerce intelligence system for consumer brands that unifies four…
Alternatives to alfred_
The closest documented capability profiles to alfred_ among enterprise operations agents tracked by Agentic Index, ordered by similarity on the same 14 point evidence the rankings use. No vendor pays for placement.
- Kyrok6.0 / 14Adds documented Prebuilt Agents, Templates & Packs
- Struction6.0 / 14Fuller documented coverage on Observability & Auditability and Model Flexibility & Routing
- Archimetis5.5 / 14Adds documented Security, Identity & Governance
- Balerion AI4.5 / 14Adds documented Security, Identity & Governance
- Humanly5.5 / 14Adds documented Prebuilt Agents, Templates & Packs
- Logility6.5 / 14Adds documented Security, Identity & Governance and Prebuilt Agents, Templates & Packs
Similarity is computed from each vendor's Agentic Index coverage score evidence, axis by axis, not from the totals. How this evidence is graded