Systems Lab
← Back to directory

Day.ai

scale your company brain

Visit Day.ai →

Day AI describes itself as a context graph and an agent control plane for go-to-market teams rather than a better CRM: it builds a shared context graph (Customer Memory) from the team's email, calendar, meetings, and Slack, and runs named AI agents with job descriptions and skills on top of it, billed per deployed agent rather than per human seat.

Vendor's own claim
CRM/authoritative state

Core capabilities

Vendor's own claim
  • Customer Memory: a context graph built from automatically permissioned ingestion of the team's email, calendar, meetings, and Slack
  • Named AI agents with a job description, skills, and a human they report to; skills run on a schedule, on company events, or on request
  • Example agent roles by tier: Sales Assistant, Meeting Notetaker, User Researcher, Onboarding Coach (Turbo); Business Development Representative, Account Executive, Sales Operator, CRM Data Entry Specialist, Sales Coach, Marketing Director (Professional); Lead Analyst, GTM Strategist, Senior Product Manager (Executive)
  • Gmail and Calendar ingestion, meeting recording and notes, and recording clips on every tier, including Free
  • Tasks, email drafts, web search, pages and documents, notifications, and lists and relationships (Turbo and above)
  • Pipeline and opportunity management, custom views, custom properties with backfill, and pipeline analytics (Professional and above)
  • Prospecting and advanced custom instructions (Executive)
  • Hosted MCP server, the open-source TypeScript day-ai-sdk, and GTM Brain, a Claude Code harness for workspace administration

API & agent surface

From official docs
Authentication
OAuth 2.0: the SDK registers its OAuth client and auto-refreshes tokens, and MCP clients are prompted to authorize in the browser on first use. MCP access inherits the Agent tier's tools, the workspace role's data access, and the tier's rate limits and quotas.
Base URL
https://day.ai/api/mcp
Rate limits
No numeric limits retrieved; the MCP page says the rate limits and quotas of the user's Agent tier apply.
Notable endpoints
  • search_objects: find contacts, organizations, opportunities, meeting recordings, and other objects with property and relationship filters
  • create_or_update_person_organization and create_or_update_opportunity: write CRM records (opportunities from Professional)
  • get_meeting_recording_context and create_meeting_recording_clip: meeting transcripts, context, and clips
  • create_email_draft and send_notification_mcp: email drafts and email or Slack notifications (Turbo and above)
  • analyze_pipeline_metrics and the CSV import tools (Professional and above)
  • search_prospects and batch create or update tools (Executive)
MCP tools exposed
search_objectscreate_or_update_person_organizationcreate_or_update_workspace_contextget_meeting_recording_contextcreate_meeting_recording_clipget_share_urlread_crm_schemaread_pageactivate_skilldeactivate_skillcreate_or_update_actioncreate_or_update_relationshipcreate_or_update_listcreate_pageupdate_pagecreate_email_draftsend_notification_mcpassistant_settingsmanage_skillswhoamicreate_or_update_opportunitycreate_or_update_custom_propertybackfill_custom_propertyanalyze_pipeline_metricscreate_import_from_filesave_import_mappingstart_importget_import_progressget_import_errorsget_imports_by_object_typeanalyze_csvread_csv_fileread_filetransform_csvcreate_viewupdate_viewconnect_slackopen_email_sharing_rulesmanage_workspace_membersbatch_create_or_update_opportunitiesbatch_create_or_update_people_organizationssearch_prospects
Machine-readable spec
No OpenAPI link

Official skills & repos

From official docs

Published by the vendor themselves — not community forks.

  • day-ai-sdk24 stars

    TypeScript SDK for using the Day AI API (MCP tools) with OAuth (MIT license); example templates: Desktop (Electron, React, Claude SDK), Desktop Agent SDK, Community Builder (Next.js), Mobile (React Native, Expo), and Vercel Cron.

  • gtm-brain11 stars

    A Claude Code harness for running a Day AI workspace like a GTM operating system: maps the GTM before signup, then plans, audits, and implements workspace configuration through the Day AI MCP server (Owner or Admin role required).

  • company-brain-evaluation1 stars

    Company-brain evaluation skill, run from Claude Code or Codex, that looks for performance and security improvements using context graph and agent control plane concepts.

  • linear-setup0 stars

    Skill to connect customer memory to product source of truth in Linear.

From the vendor's channel

Introducing Day AI: Make Every Rep Your Best Rep — Day AI

Pricing

From official docs
PlanPriceIncludedNotable
Free$0/month0 automated skill slotsGmail & Calendar ingestion, Meeting recording & notes, Meeting recording clips & context
Turbo$24/month per agent billed annually, or $30/month billed monthly2 automated skill slotsEverything in Free, Task management, Email drafts & web search
Professional$60/month per agent billed annually, or $75/month billed monthly5 automated skill slotsEverything in Turbo, Pipeline & opportunity management, Custom properties & views
Executive$200/month per agent billed annually, or $250/month billed monthly10 automated skill slotsEverything in Professional, Prospecting, Advanced custom instructions

Checked 2026-09-24 against their pricing page. Vendors change pricing without notice.

What is Day.ai

Vendor's own claim

Day AI's founder and CEO, Christopher O'Donnell, who says he led the platform at HubSpot, describes the product as "a context graph and an agent control plane" rather than a CRM with AI bolted on. Agents read and write one multiplayer context graph built from the team's conversations, and people work from the same graph, and each agent is a paid virtual worker deployed to a human colleague. The company is backed by Sequoia.

Ideal customer profile

Vendor's own claim

Sales and revenue operations teams that want to delegate repetitive CRM upkeep, prospecting research, and meeting follow-up to AI agents rather than reps or ops staff, and are comfortable paying a flat per-agent monthly fee for each virtual worker they deploy.

Who should look elsewhere

Our read

Pricing is per deployed agent, and pipeline management, custom properties, pipeline analytics, and prospecting sit on the Professional or Executive tier. The MCP page says a paid Agent tier is required to use the MCP server. The footer's API & SDK link points to the day-ai-sdk GitHub repo, whose methods call MCP tools under the hood; no separate REST API reference or OpenAPI specification was retrieved. Every subprocessor on the published list is located in the United States, and no data residency option was retrieved.

Day.ai alternatives

Our read

Systems of record for customer relationships that treat the data model as programmable, competing on API quality and flexibility rather than on breadth of built-in modules.

Category: Modern and developer-facing CRM. Membership is decided by 3 stated tests, not by category tags — see them and the full comparison.

Day.ai FAQ

Does Day AI have a public API?

Not as a standalone REST API reference. The site's API & SDK link points to the open-source day-ai-sdk, a TypeScript SDK whose convenience methods (search, createPerson, createOpportunity, sendNotification) call MCP tools on the Day AI platform; SCHEMA.md in that repo is the tool and schema reference. No OpenAPI specification was retrieved.

From official docs
Does Day AI have an MCP server?

Yes. The MCP page says a paid Agent tier is required and that the tier's tools, models, and limits apply to all MCP connections.

From official docs
Which MCP tools does Day AI expose?

The SDK's README and SCHEMA.md list 42 tools, cumulative by tier: 10 on Free (for example search_objects and get_meeting_recording_context), 10 more on Turbo, 19 more on Professional (opportunities, custom properties, imports, views), and 3 more on Executive (batch writes and search_prospects). Delete tools such as delete_contact are listed as hidden from MCP.

From official docs
Is there a free plan?

Yes, $0/month: Gmail and Calendar ingestion, meeting recording and notes, recording clips, search, share and create contacts, and workspace context notes, with no automated skill slots. Free users can search, view, and share.

From official docs
How is Day AI priced?

Per deployed AI agent, not per human seat: Free ($0), Turbo ($24/month), Professional ($60/month, most popular), and Executive ($200/month), with those prices shown for annual billing paid upfront; the Monthly toggle shows Turbo $30, Professional $75 and Executive $250 per agent per month. Volume discounts apply automatically from 21 agents. The FAQ describes annual billing as a 20% discount.

From official docs
Do I pay for every teammate?

No. Day AI charges only for the agents deployed. Workspaces with at least one Professional agent (or higher) can add colleagues at no cost, and free users can search, view, and share.

From official docs
What agent roles does Day AI offer?

Vendor-listed example agents are Sales Assistant, Meeting Notetaker, User Researcher, and Onboarding Coach (Turbo); Business Development Representative, Account Executive, Sales Operator, CRM Data Entry Specialist, Sales Coach, and Marketing Director (Professional); and Lead Analyst, GTM Strategist, and Senior Product Manager (Executive).

Vendor's own claim
Is Day AI SOC 2 compliant?

The Trust Center says Day AI's information security program includes SOC 2 Type II, encryption of data at rest and in transit, and access controls. The SOC 2 Type II report is available to customers and prospects under NDA on request.

From official docs
Does Day AI train AI models on customer data?

No, per its Data Retention & Protection Policy: Day AI does not use customer data to train, fine-tune, or otherwise improve any AI or machine learning model, and contractually requires third-party AI providers that process customer data to follow the same prohibition.

From official docs
Where does Day AI process data?

Every subprocessor on Day AI's published list, including AWS, Google, Anthropic, and OpenAI, is located in the United States. Its Data Policy commits to at least 30 days' notice before a new subprocessor processes customer data. No data residency option was retrieved.

From official docs
Is Day AI's GitHub org linked from their own site?

Yes. The site links to day-ai-sdk (API & SDK) and gtm-brain (Operations Harness), and the pricing page gives an install command for the company-brain-evaluation skill. The org's seven public repos also include linear-setup and three unrelated tooling repos (a Pulumi/Materialize packaging and forks of react-kanban and a Netlify plugin).

From official docs

Verify before you buy

  • A standalone REST API reference or OpenAPI specification separate from the SDK and MCP surface
  • Numeric MCP rate limits or quotas per Agent tier
  • A data residency option outside the United States
  • Time-boxed free trial terms for paid tiers, if any
  • A machine-readable OpenAPI specification was not included in the reviewed source set.

Need help setting it up?

This page tells you whether Day.ai fits. Cheetah builds and runs the system it goes into — enrichment, sequencing, CRM and the joins between them.

Book a call →

The directory stays free. There is nothing gated behind this.

Build my GTM engine

Tell us what you want to improve. Fedor reviews the workflow and replies by email when there is a useful fit.

Fedor reviews every request personally. If there is a useful fit, he will reply by email.