How much does Carly cost?

AI agents reported to start at 35 USD per month, with a review citing a 19 USD per month Personal plan that includes the lightweight CRM - the conflict is unresolved. **The distinctive term is that WORKFLOW STEPS THAT DO NOT CALL AN AI MODEL ARE FREE, so routing a summary into HubSpot and Slack costs nothing. Free booking pages are included. Pricing is flat or per workspace rather than per seat, which the vendor contrasts with per seat rivals that multiply across a team.**

What are the best alternatives to Carly?

Based on documented capability coverage across the Agentic Index 14 point taxonomy, the closest alternatives to Carly among enterprise operations agents are Resilinc, Vela, Apaleo, Circleback. Each is graded on the same public evidence, and no vendor pays for placement.

Does Carly support MCP and tool calling?

Full under the Agentic Index verification standard. Carly publishes public evidence of Model Context Protocol (MCP) support, tool calling, and API or SDK extensibility for connecting agents to external tools. Evidence basis: A dashboard agent builder lets customers create agents with custom instructions and configured tool access, free Zapier style workflows chain steps across connected apps, and the vendor states agents work with just about any other app alrea. Graded from public evidence only against the Agentic Index 14 point capability taxonomy. No vendor pays for placement.

Does Carly have SOC 2, SSO and role based access control?

Partial under the Agentic Index verification standard. Carly publishes some public evidence of security and identity controls, but narrower than the full set of SOC 2, single sign on (SSO) and role based access control (RBAC). Evidence basis: NOT RETRIEVED - FLAGGED, AND IT MATTERS MORE THAN USUAL HERE. An agent with its own email address acting autonomously across a user inbox, calendar and CRM is a broad trust grant, and no trust centre, attestation or data handling evidence w. Graded from public evidence only against the Agentic Index 14 point capability taxonomy. No vendor pays for placement.

Can Carly be self hosted or run in a private VPC?

Partial under the Agentic Index verification standard. Carly publishes some public evidence on deployment and data residency, but narrower than a documented self hosted or private VPC option. Evidence basis: SaaS with no application to install, running over existing email and messaging channels, which lowers deployment friction considerably but offers no self hosting and no published regional residency options. Vendor product pages 2026-08-08. Graded from public evidence only against the Agentic Index 14 point capability taxonomy. No vendor pays for placement.

Back to vendors

Carly

Q: Who is Carly for?

founders and solo operators, sales teams managing the full deal cycle, agencies coordinating many client accounts, executive assistants and recruiters

Also known as: Carly AI, usecarly, CalBot

Visit site

Enterprise operations agentindependentVerified 2026-08-08

Entry priceAI agents from 35 USD per month; a 19 USD Personal plan is also reportedFull pricing detail

Email native agent platform where every agent gets its own address, memory and tool access, and which sends and follows up on its own rather than drafting and waiting - picking up where meeting notetakers stop.

Carly is an email native agent platform, and the interface choice is the product. Users email or copy an agent to delegate work in natural language, and every agent built from the dashboard gets its own email address, its own memory, its own instructions and its own configured tool access. Agents can also be reached through Telegram, WhatsApp, Slack or text.

What separates it from the meeting and inbox tools it sits alongside is that it acts rather than drafts. The company draws the line explicitly: many AI tools draft a reply and wait for the user to press send, whereas Carly sends and follows up on its own. In appointment setting that means when someone misses a meeting it follows up the same day and offers to rebook, and when a prospect asks to move a time it reopens the negotiation, finds a new slot and updates the calendar without involving the user.

It also ships a lightweight CRM where contacts populate automatically from email and calendar activity, custom columns track whatever the user needs, and Skills update those columns from conversation context. Natural language search runs over the contact history.

The positioning against meeting notetakers is deliberate and useful for understanding the category: Carly does not capture audio at all. It integrates with capture tools and picks up where they stop, drafting and sending the follow up, updating the CRM record, creating the tasks and chasing the ones nobody actioned.

The honest caveat is sourcing. Almost every substantial description of Carly is published by Carly.

Vendor details

Canonical URL

https://www.usecarly.com

Subcategory

Email native executive assistant and workflow agent

Funding status

Private, formerly CalBot before rebranding to Carly AI. Funding detail not retrieved in this pass.

Company status

independent

Use cases & customers

Primary use cases

executing follow up work after meetings including CRM updates and task creationautonomous appointment setting, rescheduling and no show recoveryinbox and calendar management through an email addressable agentlightweight CRM populated automatically from email and calendar activitychasing action items nobody completed

Target customers

founders and solo operatorssales teams managing the full deal cycleagencies coordinating many client accountsexecutive assistants and recruiters

Deployment options

SaaS, no application to install - agents operate through existing email and messaging channels

Integrations

Reported at 260 or more integrations across 45 or more categories, though one source says 70 across 25 - the conflict is unresolved. Native CRM coverage spans HubSpot, Salesforce, Attio, Dynamics 365, Zoho and Pipedrive, alongside email (Gmail, Outlook), calendars, file storage (Google Drive, Dropbox, Box), messaging (Slack, WhatsApp, Telegram), video conferencing (Zoom, Google Meet, Microsoft Teams) and accounting (QuickBooks, Xero, FreshBooks). Each agent is given its own configured tool access rather than inheriting a global connection set.

In practice

A prospect misses a booked meeting and the agent follows up the same day offering to rebook, rather than the thread dying.

A meeting ends and the agent drafts and sends the follow up, updates the CRM record, creates the tasks and later chases the ones nobody actioned.

A team copies an agent on an email thread the way they would a colleague, and its work appears in the same thread everyone can already see.

Sources & related URLs

Research sources

https://aiagentstore.ai/ai-agent/carly-ai https://www.usecarly.com/blog/best-ai-agent-2026/ https://www.usecarly.com/blog/ai-agent-for-appointment-setting/ https://www.usecarly.com/blog/best-ai-crm-tools/ https://slashai.app/reviews/carly-ai-review/ https://tryanai.com/tool/carly-ai

Agentic Index coverage score

8.5 / 14 capabilities · 61%

Integrations & Tool CallingReported at 260 plus integrations across 45 plus categories, though another source says 70 across 25 - conflict unresolved. Native CRM coverage spans HubSpot, Salesforce, Attio, Dynamics 365, Zoho and Pipedrive, plus email, calendars, file storage, messaging, video conferencing and accounting. Notably each agent receives its own CONFIGURED TOOL ACCESS rather than inheriting a global connection set. Vendor product pages 2026-08-08	Full
Workflow OrchestrationExecutes multi step tasks across email, calendar, CRM and other tools from high level natural language instructions, with custom agents built from a dashboard each carrying its own instructions and tool access. An independent directory assessment specifically credits high autonomy on this basis. Appointment negotiation runs end to end including rescheduling without user involvement. Independent directory assessment and vendor product pages 2026-08-08	Full
Knowledge Grounding & RAGA built in lightweight CRM populates contacts automatically from email and calendar activity and supports natural language search over that history, with contact enrichment and web research available to agents, but no retrieval, vector or knowledge grounding layer is documented as a product surface. Vendor product pages 2026-08-08	Partial
Human Oversight & GuardrailsDELIBERATELY AUTONOMY FORWARD, and the vendor says so: many AI tools draft the reply and wait for you to hit send, whereas Carly SENDS AND FOLLOWS UP ON ITS OWN. Each agent has configured access permissions and rules which bound its reach, but no approval gate, review step or confirmation mechanism was documented. Same posture as abnormal-ai - a design choice rather than an omission, but the axis measures oversight and the oversight is thin. Vendor product pages 2026-08-08	Partial
Security, Identity & GovernanceNOT RETRIEVED - FLAGGED, AND IT MATTERS MORE THAN USUAL HERE. An agent with its own email address acting autonomously across a user inbox, calendar and CRM is a broad trust grant, and no trust centre, attestation or data handling evidence was retrieved. Per agent access permissions are documented as a control. Check usecarly.com security pages and correct. Vendor and third party sources 2026-08-08	Partial
Observability & AuditabilityAn unusual structural property rather than a feature: because each agent has its own email address and works in the thread, ITS ACTIONS ARE VISIBLE IN THE INBOX BY CONSTRUCTION to everyone on the thread - a low tech cousin of assigning agents work items in a shared tracker. A dashboard manages agents. No agent decision log, reasoning trace or audit surface was documented. Vendor product pages 2026-08-08	Partial
Memory & State PersistenceEACH AGENT GETS ITS OWN MEMORY as a stated platform property, with the vendor describing keeping the agent running after the user closes their laptop and SAVING ITS MEMORY. Skills additionally update CRM columns automatically from conversation context, so learned state feeds back into structured data. Per agent memory rather than a shared data store. Vendor product pages and directory listing 2026-08-08	Full
Deployment & Data ResidencySaaS with no application to install, running over existing email and messaging channels, which lowers deployment friction considerably but offers no self hosting and no published regional residency options. Vendor product pages 2026-08-08	Partial
Prebuilt Agents, Templates & PacksAn agent builder ships documented patterns customers assemble directly - a sales pipeline agent logging interactions and updating deal stages, a client intake agent qualifying leads and booking meetings, a follow up agent checking in with dormant contacts - alongside free booking pages and free Zapier style workflows. Vendor product pages 2026-08-08	Full
Triggers & Channel CoverageGENUINELY EVENT DRIVEN, not merely scheduled: a missed meeting produces a same day follow up offering to rebook, and a reschedule request reopens the negotiation automatically. Scheduled check ins on dormant contacts are also supported, so both modes exist. Reachable by email, text, Telegram, WhatsApp and Slack, with each agent addressable at its own email address. Vendor product pages 2026-08-08	Full
Model Flexibility & RoutingNo model selection, provider choice, routing configuration or bring your own model capability was documented in any source retrieved. Honest absence. Vendor and third party sources 2026-08-08	Unable to verify
APIs, SDKs & MCP ExtensibilityA dashboard agent builder lets customers create agents with custom instructions and configured tool access, free Zapier style workflows chain steps across connected apps, and the vendor states agents work with just about any other app already in use beyond the native integration list. No public API reference, SDK or MCP support was retrieved. Vendor product pages 2026-08-08	Full
Testing, Debugging & OptimizationThe vendor mentions testing an agent in the browser before adding channels, but no agent evaluation, scoring, simulation or regression capability was documented - notable for a product whose agents send email autonomously on the user behalf. Honest absence. Vendor product pages 2026-08-08	Unable to verify
Browser & Computer UseAgents act through integrations and email rather than by driving a browser, and the vendor positions API based integration explicitly as the alternative to browser automation for recurring work. No computer use capability exists. Vendor blog and product pages 2026-08-08	Unable to verify

The Agentic Index coverage score grades every vendor Full, Partial or Unable to verify 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

Pricing

AI agents from 35 USD per month; a 19 USD Personal plan is also reported

flat or per workspace subscription; non AI workflow steps free

Free tierTrial available

Included quota

Workflow steps that do not invoke an AI model are free and unmetered. Agent quotas at the paid tiers were not retrieved.

What is public

The free non AI workflow steps, free booking pages and a 35 USD per month agent entry point are consistently reported. Plan detail is not.

Billing mechanics

Self serve subscription. Non AI workflow steps are free; AI agents are the paid layer. Flat or per workspace rather than per seat.

Cost watchouts

The pricing structure is unusually clean, so the watchouts are operational rather than financial. **The real exposure is trust rather than cost: an agent with its own email address sending and following up autonomously across inbox, calendar and CRM can act at volume before anyone reviews it, and no approval gate or evaluation surface was documented.** Note also that source figures conflict (35 USD versus a 19 USD Personal plan) and that integration counts are reported as both 260 plus and 70 plus, so verify the plan and the connector coverage against first party pages before committing.

Variable cost rationale

Graded low because the subscription is flat or per workspace rather than per seat, non AI workflow steps are explicitly free and unmetered, and there is no agent consumption unit, token pass through or per action meter of any kind. A buyer can forecast cost from the plan alone. The residual uncertainty is which plan applies, since sources conflict between 19 and 35 USD, not how consumption behaves.

Additional watchouts

Verify the plan and connector coverage first party - both pricing and integration counts conflict across sources. And weigh the trust grant rather than the price: this agent sends on its own.

Overage / add-ons

Not retrieved.

Sales call required

No, self serve available

Free / trial

Free Zapier style workflows and free booking pages; AI agents are the paid layer

Lowest paid plan

AI agents from 35 USD per month, with a 19 USD Personal plan also reported

Commercial notes

**METER THE INTELLIGENCE, NOT THE PLUMBING IS THE MOST RATIONAL PRICING STRUCTURE ENCOUNTERED IN THIS SWEEP, and it is worth setting against the alternatives: seven vendors have invented named agent consumption units that meter everything indiscriminately and none publishes a rate per unit of real work, while the durable execution pocket produces a paradox where finer grained recovery means more billable step executions. Carly charges only where a model is actually invoked. WATCH WHETHER OTHERS ADOPT IT.** SOURCE WARNING: almost every substantial description of Carly is published by Carly, which runs a systematic self ranking content operation placing itself first across many category roundups. Product claims here are treated as vendor claims; its rankings were used for nothing.

Key ambiguities

Whether the entry point is 19 or 35 USD per month and what separates those plans, and whether AI agent usage carries any volume limit above the subscription. Neither was retrievable from independent sources.

Missing data

Plan structure and the difference between the reported 19 and 35 USD tiers, agent volume limits, and any enterprise tier.

Agentic Index verified 2026-08-08

Official pricing page

Related vendors

Alternatives to Carly

The closest documented capability profiles to Carly 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.

Resilinc9.0 / 14Adds documented Testing, Debugging & Optimization
Vela8.0 / 14Adds documented Testing, Debugging & Optimization
Apaleo6.5 / 14A lighter documented profile than Carly
Circleback8.5 / 14Fuller documented coverage on Knowledge Grounding & RAG and Security, Identity & Governance
Orbio7.5 / 14Fuller documented coverage on Security, Identity & Governance
Rubick.ai6.5 / 14A lighter documented profile than Carly

Similarity is computed from each vendor's Agentic Index coverage score evidence, axis by axis, not from the totals. How this evidence is graded

Compare Carly with other vendors →Compare all enterprise operations agents →See where Carly ranks →

Data confidence: medium