---
segment: human-handoff
segment_kind: useCase
question: "What is the best AI customer service agent for handing customers over to a person?"
page_url: https://ai-agent.review/best-ai-agent-for-customer-service/human-handoff
record_url: https://ai-agent.review/best-ai-agent-for-customer-service/human-handoff.md
tested_performance: not-yet-tested
---

# What is the best AI customer service agent for handing customers over to a person?

No agent is the best AI customer service agent for handing customers over to a person, because nobody has tested them against each other. The best-evidenced are Talkdesk, Forethought and NiCE CXone: an external source, not the vendor, reports each of them handling handing customers over to a person. 21 of the 22 agents reviewed have any published evidence for handing customers over to a person; for the rest it is unknown, which means nobody has published it, not that the product cannot do it. Nothing here has been tested by this portal, so treat every name below as a shortlist to validate, not a winner.

## How these groups are formed

1. Agents are grouped by what the cited sources establish for this segment, not scored against each other.
2. An outside source reporting the product in this segment comes first, then the vendor's own documentation, then a documented limitation, then silence.
3. Within a group, the agent with more of the seven customer workflows reported by an outside source comes first, then the one with more vendor-documented workflows; only where both counts tie is the order alphabetical.
4. No agent is dropped: the ones with nothing published are named too.

Groups and the order inside them are derived from the cited claims at build time and describe the evidence, not the products. Nothing here has been tested, so a higher group or a higher place inside one means more has been established about an agent, not that it performs better. An agent can be well suited to you and sit in the last group simply because nobody has written about it. Published review-site ratings do not order these groups either: where a record carries one it is cited on the profile, but those ratings score whole products against each other — one vendor's bundled chat widget against another's full help desk — and none of them measures the AI agent on its own.

## An outside source reports it serving this segment

A customer, analyst or journalist with no stake in the sale describes the product working here. This is the strongest evidence the portal holds, and it is still somebody else's deployment, not yours.

Ordered by customer workflows with evidence: externally reported first, then vendor-documented, A–Z on a tie. Both counts are shown with every agent. A higher place means more has been established, not that the agent performs better.

| Agent | Vendor | Evidence for this segment | What puts it here | Customer workflows with evidence |
| --- | --- | --- | --- | --- |
| [Talkdesk](https://ai-agent.review/agents/talkdesk.md) | Talkdesk | External report | [Talkdesk Launches AI Agents for Retail, Goes All In on Agentic AI](https://www.cxtoday.com/contact-center/talkdesk-launches-ai-agents-for-retail-continues-to-prioritize-agentic-ai/) | 3 externally reported, 2 vendor-documented, of 7 |
| [Forethought](https://ai-agent.review/agents/forethought.md) | Forethought (Zendesk) | External report | [Zendesk announces deal to acquire customer support AI startup Forethought](https://siliconangle.com/2026/03/11/zendesk-announces-deal-acquire-customer-support-ai-startup-forethought/) | 3 externally reported, 0 vendor-documented, of 7 |
| [NiCE CXone](https://ai-agent.review/agents/nice-cxone.md) | NiCE | External report | [Customer Journey Orchestration Is Getting Serious. Is NiCE Agentic AI Architecture Already Ahead?](https://www.cxtoday.com/customer-engagement-journey-orchestration/nice-customer-journey-orchestration/) | 3 externally reported, 0 vendor-documented, of 7 |
| [Crescendo](https://ai-agent.review/agents/crescendo.md) | Crescendo | External report | [Alorica and Crescendo Partner to Bring AI into Live Enterprise Service Channels](https://www.cxtoday.com/ai-automation-in-cx/alorica-and-crescendo-partner-to-bring-ai-into-live-enterprise-service-channels/) | 2 externally reported, 1 vendor-documented, of 7 |
| [Lyro](https://ai-agent.review/agents/lyro.md) | Tidio | External report | [Tidio Live Chat and AI Chatbots reviews (Shopify App Store)](https://apps.shopify.com/tidio-chat/reviews) | 2 externally reported, 1 vendor-documented, of 7 |
| [Five9](https://ai-agent.review/agents/five9.md) | Five9 | External report | [Five9 Co-Innovates with ServiceNow, Unveils Its AI Agents](https://www.cxtoday.com/contact-center/five9-co-innovates-with-servicenow-unveils-its-ai-agents/) | 2 externally reported, 0 vendor-documented, of 7 |
| [Decagon](https://ai-agent.review/agents/decagon.md) | Decagon | External report | [Krishan Y., G2 review syndicated by AWS Marketplace](https://aws.amazon.com/marketplace/reviews/reviews-list/prodview-pnbnafhjefhhy) | 1 externally reported, 6 vendor-documented, of 7 |
| [Zendesk AI Agents](https://ai-agent.review/agents/zendesk-ai-agents.md) | Zendesk | External report | [Kārlis Stenders: Advanced AI Agent escalation form (Zendesk Community)](https://community.zendesk.com/chat-messaging-13/hot-to-fix-the-advanced-ai-agent-human-escalation-form-description-field-22417) | 1 externally reported, 6 vendor-documented, of 7 |
| [Kore.ai](https://ai-agent.review/agents/kore-ai.md) | Kore.ai | External report | [Kore.ai Makes Its Third Wave Play, Multi-Agent Orchestration For CX](https://www.cxtoday.com/ai-automation-in-cx/kore-ai-makes-its-third-wave-play-multi-agent-orchestration-for-cx/) | 1 externally reported, 1 vendor-documented, of 7 |
| [NiCE Cognigy](https://ai-agent.review/agents/nice-cognigy.md) | NiCE | External report | [Why Is NiCE Spending Almost $1BN on Cognigy? A Deep Dive](https://www.cxtoday.com/contact-center/why-is-nice-spending-almost-1bn-on-cognigy-a-deep-dive/) | 1 externally reported, 1 vendor-documented, of 7 |
| [Sprinklr](https://ai-agent.review/agents/sprinklr.md) | Sprinklr | External report | [Sprinklr Service and the Shift from AI Explainability to AI Assurance](https://www.cxtoday.com/contact-center/sprinklr-service-and-ai-assurance/) | 1 externally reported, 1 vendor-documented, of 7 |

## The vendor documents this segment, and nobody else has written about it

The vendor says it serves this segment. That is a reasonable place to start a shortlist and it is not confirmation, because no independent source has reported on it either way.

Ordered by customer workflows with evidence: externally reported first, then vendor-documented, A–Z on a tie. Both counts are shown with every agent. A higher place means more has been established, not that the agent performs better.

| Agent | Vendor | Evidence for this segment | What puts it here | Customer workflows with evidence |
| --- | --- | --- | --- | --- |
| [Ada](https://ai-agent.review/agents/ada.md) | Ada | Vendor-documented | [Handoff management (Ada docs)](https://docs.ada.cx/docs/handoffs/handoff-management) | 0 externally reported, 7 vendor-documented, of 7 |
| [Agentforce](https://ai-agent.review/agents/agentforce.md) | Salesforce | Vendor-documented | [Agentforce Service Agent (Salesforce Help)](https://help.salesforce.com/s/articleView?id=service.service_agent_overview.htm&language=en_US&type=5) | 0 externally reported, 7 vendor-documented, of 7 |
| [Fin](https://ai-agent.review/agents/fin.md) | Intercom | Vendor-documented | [Manage Fin AI Agent's escalation guidance and rules (Intercom Help)](https://www.intercom.com/help/en/articles/12396892-manage-fin-ai-agent-s-escalation-guidance-and-rules) | 0 externally reported, 7 vendor-documented, of 7 |
| [Sierra](https://ai-agent.review/agents/sierra.md) | Sierra | Vendor-documented | [Agent SDK (Sierra)](https://sierra.ai/product/agent-sdk) | 0 externally reported, 7 vendor-documented, of 7 |
| [Gorgias AI Agent](https://ai-agent.review/agents/gorgias.md) | Gorgias | Vendor-documented | [Customize how AI Agent hands over to your team](https://docs.gorgias.com/en-US/customize-how-ai-agent-hands-over-to-your-team-6008591) | 0 externally reported, 5 vendor-documented, of 7 |
| [Breeze Customer Agent](https://ai-agent.review/agents/hubspot-breeze.md) | HubSpot | Vendor-documented | [Understand the customer agent — HubSpot Knowledge Base](https://knowledge.hubspot.com/customer-agent/understand-the-customer-agent) | 0 externally reported, 2 vendor-documented, of 7 |
| [Genesys Cloud CX](https://ai-agent.review/agents/genesys-cloud-cx.md) | Genesys | Vendor-documented | [Virtual Agents - Genesys](https://www.genesys.com/capabilities/virtual-agents) | 0 externally reported, 2 vendor-documented, of 7 |
| [LivePerson](https://ai-agent.review/agents/liveperson.md) | LivePerson | Vendor-documented | [AI Chatbots — LivePerson](https://www.liveperson.com/products/ai-chatbots/) | 0 externally reported, 2 vendor-documented, of 7 |
| [Yellow.ai](https://ai-agent.review/agents/yellow-ai.md) | Yellow.ai | Vendor-documented | [Escalation (Yellow.ai docs)](https://docs.yellow.ai/docs/nexus/build/tools/escalation) | 0 externally reported, 2 vendor-documented, of 7 |

## A limitation is documented for this segment

A cited source describes a limit or a bad experience here. It sits below vendor documentation because it is not evidence of fit, and above the agents nobody has written about because somebody at least looked. Read the condition: it may not apply to you at all.

| Agent | Vendor | Evidence for this segment | What puts it here | Customer workflows with evidence |
| --- | --- | --- | --- | --- |
| [PolyAI](https://ai-agent.review/agents/polyai.md) | PolyAI | Caveat documented | [PolyAI Docs — Voice channel: Handoffs](https://docs.poly.ai/voice-channel/handoffs.md) | 2 externally reported, 1 vendor-documented, of 7 |

## Nothing published for this segment

No reviewed source places these agents here, by the vendor's own documentation or otherwise. That is a gap in the public record, not a finding against the products, and several of them may suit you.

- [Freddy AI Agent](https://ai-agent.review/agents/freddy-ai.md): 0 externally reported, 0 vendor-documented, of 7

## Every claim behind these groups

### Talkdesk — External report

- Talkdesk states Autopilot seamlessly redirects customers to a human agent when it detects a complex case requiring human expertise. _(Vendor-reported, checked 20 Sept 2026; source: [Talkdesk Autopilot — Self-Service Experience](https://www.talkdesk.com/cloud-contact-center/self-service-experience/autopilot/); claim id: `talkdesk-fit-handoff-vendor`)_
- Trade coverage of Talkdesk AI Agents for Retail reports the agents support routing to in-store specialists, escalating from AI to a human when necessary. _(Independent report, published 10 Jan 2025, checked 20 Sept 2026; source: [Talkdesk Launches AI Agents for Retail, Goes All In on Agentic AI](https://www.cxtoday.com/contact-center/talkdesk-launches-ai-agents-for-retail-continues-to-prioritize-agentic-ai/); claim id: `talkdesk-fit-handoff-indep`)_

### Forethought — External report

- The vendor states Solve hands conversations off to human agents when the AI cannot resolve an issue or the customer asks for a person, with the Triage agent classifying and routing the ticket to the appropriate agent with full context. _(Vendor-reported, checked 20 Sept 2026; source: [Forethought Solve](https://forethought.ai/solve/); claim id: `forethought-handoff-vendor`)_
- SiliconANGLE reports that the system determines whether an issue can be resolved automatically or requires escalation to a human agent. _(Independent report, published 11 Mar 2026, checked 20 Sept 2026; source: [Zendesk announces deal to acquire customer support AI startup Forethought](https://siliconangle.com/2026/03/11/zendesk-announces-deal-acquire-customer-support-ai-startup-forethought/); claim id: `forethought-handoff-independent`)_

> All three fit verdicts in this record rest on the same single independent article about the acquisition, so they stand or fall together rather than representing independent corroboration.

### NiCE CXone — External report

- NiCE states its self-service AI agents hand over to a human agent when the conversation hits escalation parameters, supporting a smooth handover. _(Vendor-reported, checked 20 Sept 2026; source: [AI Agents for Self-Service](https://www.nice.com/products/ai-agents-for-self-service); claim id: `nice-cxone-fit-handoff-vendor`)_
- CX Today describes NiCE's agentic AI as preserving context across channels and across handoffs between AI agents, human agents and back-office systems. _(Independent report, published 23 Jul 2026, checked 20 Sept 2026; source: [Customer Journey Orchestration Is Getting Serious. Is NiCE Agentic AI Architecture Already Ahead?](https://www.cxtoday.com/customer-engagement-journey-orchestration/nice-customer-journey-orchestration/); claim id: `nice-cxone-fit-handoff-independent`)_

> The independent article describes the architecture rather than verifying handoff quality in production, and separately questions how much work still reaches human queues.

### Crescendo — External report

- Crescendo states that when Concierge escalates it "passes along the entire conversation — customer history, issue context, recommended next action" so the specialist "starts at the answer". _(Vendor-reported, checked 20 Sept 2026; source: [Crescendo — Concierge Agent](https://www.crescendo.ai/agents/concierge); claim id: `crescendo-fit-handoff-vendor`)_
- CX Today reported on the Alorica–Crescendo partnership that AI "handles the heavy lifting upfront" and that "when escalation is needed, that full context is passed seamlessly to a human agent", with Alorica's 100,000+ employees across 16 countries available to augment Crescendo deployments. _(Independent report, published 26 Jun 2026, checked 20 Sept 2026; source: [Alorica and Crescendo Partner to Bring AI into Live Enterprise Service Channels](https://www.cxtoday.com/ai-automation-in-cx/alorica-and-crescendo-partner-to-bring-ai-into-live-enterprise-service-channels/); claim id: `crescendo-fit-handoff-ind-alorica`)_

> Handoff here is not only a software escalation: the humans on the receiving end can be supplied by Crescendo itself (via the acquired PartnerHero workforce) or by partner BPOs, so a buyer may be purchasing the escalation tier as well as the AI. The independent article's description of the handoff is sourced to partner and vendor executives quoted by the reporter.

### Lyro — External report

- Tidio states users can set flexible handoff and escalation rules and that Lyro creates a ticket in the connected support solution so a human agent can answer. _(Vendor-reported, checked 20 Sept 2026; source: [Lyro AI Agent (Tidio)](https://www.tidio.com/lyro/); claim id: `lyro-fit-handoff-vendor`)_
- A Shopify merchant described the agent as able to answer repetitive questions automatically and escalate questions when needed. _(Independent report, published 21 Aug 2026, checked 20 Sept 2026; source: [Tidio Live Chat and AI Chatbots reviews (Shopify App Store)](https://apps.shopify.com/tidio-chat/reviews); claim id: `lyro-fit-handoff-ind-1`)_

### Five9 — External report

- Five9 states that "when escalation is needed, full conversation context transfers instantly to a live agent, so customers never start over." _(Vendor-reported, checked 20 Sept 2026; source: [Five9 AI Agents](https://www.five9.com/products/capabilities/ai-agents); claim id: `five9-handoff-context`)_
- CX Today reported that with Five9 AI Agents, "contact centers can add guardrails to ensure these conversations still pass through to a live agent" when needed. _(Independent report, published 13 Nov 2024, checked 20 Sept 2026; source: [Five9 Co-Innovates with ServiceNow, Unveils Its AI Agents](https://www.cxtoday.com/contact-center/five9-co-innovates-with-servicenow-unveils-its-ai-agents/); claim id: `five9-handoff-guardrails`)_

### Decagon — External report

- Decagon documents live-chat escalation, email handoff and voice transfer to human agents. _(Vendor-reported, checked 20 Sept 2026; source: [Integrations (Decagon)](https://decagon.ai/product/integrations); claim id: `decagon-fit-handoff-vendor`)_
- Krishan Y.’s August 12, 2026 review says their team can step in when human attention is needed. This corroborates takeover in one self-reported deployment, without measuring routing accuracy or waiting time. _(Independent report, published 12 Aug 2026, checked 20 Sept 2026; source: [Krishan Y., G2 review syndicated by AWS Marketplace](https://aws.amazon.com/marketplace/reviews/reviews-list/prodview-pnbnafhjefhhy); claim id: `decagon-human-handoff-g2-krishan`)_

> Illustrative evaluation: request a human after an unsuccessful troubleshooting step. Check queue routing, transcript transfer, repeated questions and the offline fallback.

### Zendesk AI Agents — External report

- Zendesk states that its AI agents can escalate to human teams with conversation context. _(Vendor-reported, checked 20 Sept 2026; source: [Zendesk AI Agents product overview](https://www.zendesk.com/service/ai/ai-agents/); claim id: `zendesk-fit-handoff-vendor`)_
- Kārlis Stenders describes collecting an issue description conversationally so it appears in the transcript during an Advanced AI Agent messaging handoff. The same post reports form limitations and corrects a custom workaround; it does not quantify reliability. _(Independent report, published 4 Sept 2026, checked 20 Sept 2026; source: [Kārlis Stenders: Advanced AI Agent escalation form (Zendesk Community)](https://community.zendesk.com/chat-messaging-13/hot-to-fix-the-advanced-ai-agent-human-escalation-form-description-field-22417); claim id: `zendesk-human-handoff-transcript-carryover`)_

> Illustrative scenario: a customer asks for a human after troubleshooting. Verify contact capture, transcript transfer, destination queue, and the ability to reply on the intended channel. The cited community workaround shows one implementation path, not a general handoff success rate.

### Kore.ai — External report

- Kore.ai documents escalation to a human agent that passes session variables and conversation history via a context_for_human block, converting routing metadata to SIP headers on voice channels and forwarding it through the provider's API on digital channels. _(Vendor-reported, checked 20 Sept 2026; source: [Escalation to human agent - Kore.ai Agent Platform documentation](https://koreai.mintlify.app/agent-platform/escalate.md); claim id: `kore-ai-fit-handoff-vendor`)_
- CX Today reported that Kore.ai's agent platform supports orchestration patterns including handoff and escalation across channels, and quoted the design position that handing off to a human is part of the journey rather than a failure state. _(Independent report, published 21 May 2026, checked 20 Sept 2026; source: [Kore.ai Makes Its Third Wave Play, Multi-Agent Orchestration For CX](https://www.cxtoday.com/ai-automation-in-cx/kore-ai-makes-its-third-wave-play-multi-agent-orchestration-for-cx/); claim id: `kore-ai-fit-handoff-cxtoday`)_

> The same CX Today piece flags context preservation across agent handoffs as the thing buyers should verify; Kore.ai's own documentation lists missing conversation_history and incomplete on_human_complete handlers as implementation failure points.

### NiCE Cognigy — External report

- Cognigy documents ten named handover providers for escalating conversations to human agents, including NiCE CXone, Genesys Cloud, Salesforce Service Cloud, Salesforce MIAW, 8x8, RingCentral Engage, Chatwoot, storm and Cognigy Live Agent. _(Vendor-reported, checked 20 Sept 2026; source: [Handover Provider Reference - Overview \| Cognigy Documentation](https://docs.cognigy.com/ai/escalate/handover-reference/overview.md); claim id: `nice-cognigy-fit-handoff-vendor`)_
- CX Today reports that Cognigy grew via partnerships with contact centre providers including Avaya, Genesys and RingCentral before developing into a standalone platform. _(Independent report, published 1 Aug 2025, checked 20 Sept 2026; source: [Why Is NiCE Spending Almost $1BN on Cognigy? A Deep Dive](https://www.cxtoday.com/contact-center/why-is-nice-spending-almost-1bn-on-cognigy-a-deep-dive/); claim id: `nice-cognigy-fit-handoff-ind1`)_

> Cognigy's documentation states its native Cognigy Live Agent handover provider covers chat conversations, and that real-time queue notifications are available only with Cognigy Live Agent.

### Sprinklr — External report

- Sprinklr states its AI Agents "Respond with empathy and precision. Intelligently escalate to human agents when needed." _(Vendor-reported, checked 20 Sept 2026; source: [Sprinklr AI Agents](https://www.sprinklr.com/products/platform/ai-agents/); claim id: `sprinklr-handoff-vendor`)_
- CX Today reports the Sprinklr deployment at telecom operator Umniah was cited with a 53% reduction in handovers to human agents. _(Independent report, published 20 Sept 2026, checked 20 Sept 2026; source: [Sprinklr Service and the Shift from AI Explainability to AI Assurance](https://www.cxtoday.com/contact-center/sprinklr-service-and-ai-assurance/); claim id: `sprinklr-handoff-umniah`)_

> The handover figure originates from a named customer deployment reported by the publication; Sprinklr has not published a methodology for it.

### Ada — Vendor-documented

- Ada documents live transfers, follow-up tickets, off-hours paths and error fallbacks across its handoff setup. _(Vendor-reported, checked 20 Sept 2026; source: [Handoff management (Ada docs)](https://docs.ada.cx/docs/handoffs/handoff-management); claim id: `ada-fit-handoff-vendor`)_

> Illustrative scenario, not a test: a customer requests a person after an unsuccessful answer. Verify that the receiving team gets the conversation and that, outside working hours, the customer receives a clear follow-up path.

### Agentforce — Vendor-documented

- Service Agents use configured Omni-Channel Flow routing to transfer support requests to service representatives or other destinations. _(Vendor-reported, checked 20 Sept 2026; source: [Agentforce Service Agent (Salesforce Help)](https://help.salesforce.com/s/articleView?id=service.service_agent_overview.htm&language=en_US&type=5); claim id: `agentforce-fit-handoff-vendor`)_

> Illustrative scenario, not a test: a customer asks to speak to a person. Confirm the destination queue is staffed, the conversation reaches it and the customer receives an off-hours path when nobody is available.

### Fin — Vendor-documented

- Fin uses configured workflows to route escalations. Intercom states that no escalation offer is shown when no human routing target exists. _(Vendor-reported, checked 20 Sept 2026; source: [Manage Fin AI Agent's escalation guidance and rules (Intercom Help)](https://www.intercom.com/help/en/articles/12396892-manage-fin-ai-agent-s-escalation-guidance-and-rules); claim id: `fin-handoff-vendor`)_

> Illustrative scenario: a customer asks for a person after an unsuccessful answer. Verify the destination inbox, transcript transfer, and after-hours response. A handoff event alone does not show how quickly a human replies.

### Sierra — Vendor-documented

- Sierra’s SDK page documents routing an escalated conversation to a team member with a generated summary. It does not establish how quickly a human answers in a particular contact center. _(Vendor-reported, checked 20 Sept 2026; source: [Agent SDK (Sierra)](https://sierra.ai/product/agent-sdk); claim id: `sierra-handoff-sdk`)_

> Illustrative evaluation: request a person during a failed account update. Check queue routing, transcript and summary delivery, plus what the customer sees when the team is offline.

### Gorgias AI Agent — Vendor-documented

- Gorgias documents built-in handover signals (customer anger, explicit requests for a human, sensitive topics such as self-harm, violence, legal threats or financial account requests), configurable handover topics, and a pre-send draft review where "If it doesn't have enough knowledge to answer reliably...it hands over rather than guessing." _(Vendor-reported, checked 3 Oct 2026; source: [Customize how AI Agent hands over to your team](https://docs.gorgias.com/en-US/customize-how-ai-agent-hands-over-to-your-team-6008591); claim id: `gorgias-fit-handoff-vendor`)_

> Handover configuration differs by channel: email and SMS expose customer-notification settings, while chat adds confirmation prompts, wait-time messaging and email capture.

### Breeze Customer Agent — Vendor-documented

- HubSpot documents that the agent transfers a conversation to a human when its confidence is low, when escalation rules are met, when a visitor explicitly asks for a human, or when a configured handoff phrase is triggered. _(Vendor-reported, checked 20 Sept 2026; source: [Understand the customer agent — HubSpot Knowledge Base](https://knowledge.hubspot.com/customer-agent/understand-the-customer-agent); claim id: `hubspot-breeze-fit-handoff-vendor`)_

> No independent evaluation of handoff quality or escalation accuracy was found.

### Genesys Cloud CX — Vendor-documented

- Genesys states its Agentic Virtual Agent provides seamless, contextual handoffs with full context when humans step in. _(Vendor-reported, checked 20 Sept 2026; source: [Virtual Agents - Genesys](https://www.genesys.com/capabilities/virtual-agents); claim id: `genesys-fit-handoff`)_

> Marketing-page claim; the specific fields or transcript passed to the human agent are not documented on the sources reviewed.

### LivePerson — Vendor-documented

- LivePerson states its AI chatbots can 'seamlessly transfer complex customer queries to live agents' and describes a 'Bot Human Tango' model in which both humans and bots contribute to a conversation with handoff from one type of agent to the other. _(Vendor-reported, checked 20 Sept 2026; source: [AI Chatbots — LivePerson](https://www.liveperson.com/products/ai-chatbots/); claim id: `liveperson-fit-handoff-vendor`)_

> No independent evaluation of handoff quality in production was located, so this area remains unverified.

### Yellow.ai — Vendor-documented

- Yellow.ai documents two escalation tools: Escalate to Agent, which routes chat into Inbox queues with round-robin, least-busy or specific-group assignment and passes a conversation summary to the human agent, and Transfer Call, which forwards voice calls over PSTN or SIP with custom headers for context. _(Vendor-reported, checked 20 Sept 2026; source: [Escalation (Yellow.ai docs)](https://docs.yellow.ai/docs/nexus/build/tools/escalation); claim id: `yellow-ai-fit-handoff-vendor`)_

> The escalation documentation notes routing rules cannot name a tool, so escalation must be triggered from the agent's instructions rather than routing logic, and warns that malformed SIP headers can drop calls during voice transfer. Both are authoring constraints with documented workarounds rather than capability gaps.

### PolyAI — Caveat documented

- PolyAI documents voice handoff via SIP REFER, SIP INVITE or SIP BYE with escalation context passed through custom X- SIP headers, a Handoff State API endpoint and a Conversations API transcript, plus chat handoff integrations for Salesforce, Zendesk, NICE CXone, Amazon Connect, Webex and Five9. _(Vendor-reported, checked 20 Sept 2026; source: [PolyAI Docs — Voice channel: Handoffs](https://docs.poly.ai/voice-channel/handoffs.md); claim id: `polyai-handoff-vendor`)_
- PolyAI's handoff documentation states that with the default SIP REFER method "the utterance may not play before the transfer completes because the REFER fires at the same time as function execution", and that inbound SIP voicemail detection on the destination side is "not available by default" for SIP REFER, BYE or INVITE transfers. _(Vendor-reported, checked 20 Sept 2026; source: [PolyAI Docs — Voice channel: Handoffs](https://docs.poly.ai/voice-channel/handoffs.md); claim id: `polyai-handoff-gap`)_

Method: https://ai-agent.review/methodology#shortlists