Aircover.ai

A named catalog spanning five classes, with authenticated action documented rather than inferred. Web conferencing: Zoom, Microsoft Teams, Google Meet and Webex, each with full real-time support. CRM: Salesforce with bi-directional sync, HubSpot with automatic field mapping, ServiceNow with workflow automation.

Sales enablement: Highspot for content delivery, Salesloft and Nooks as dialers, Outreach for call coaching. Communication and cloud: Google Calendar and Outlook for calendar sync, Twilio and Vonage for contact center, ServiceNow for notifications. Plus Google Drive for content import. The agents write as well as read, syncing parsed call data into custom CRM fields and triggering downstream workflows.

Sourceaircover.ai/partners-integrationsread 2026-09-01

Aircover runs multiple agents with a named orchestration surface. It ships an Agent Orchestrator as its own product alongside a set of specialist agents, describing them as a team of digital workers each expert in one part of the deal and deployed before, during and after every call, with the site's own strapline being a team of specialist agents with you in every conversation. Work chains across them: Call Prep researches the account ahead of the meeting, Live Coaching and Objection Handling run in the conversation, and AI Notes, AI CRM Sync and follow-up automation run afterwards.

Sourceaircover.ai/agent-orchestrator and aircover.ai/agentsread 2026-09-01

In call AI is connected to the company's document repository to surface the right content live, and the MCP scope names read access to indexed documents through an ask_documents tool, a maintained index over the customer's own content.

Sourceaircover.ai/developersread 2026-10-01

No approval gate is documented, though the agents act. Aircover's agents write bi-directionally to Salesforce, map fields into HubSpot, trigger ServiceNow workflow automation, and the sales page describes creating and sending quotes and automating follow-ups; the Procurement agent works the buying process.

Nothing across the agents page, the platform pages, the integrations page or the trust center documents an approval queue, a review step before an agent writes or sends, an autonomy setting or permission scoping over what an agent may do. The configurable transcription notice options on the trust center are a compliance disclosure to call participants, not an oversight control over the agent.

Sourceaircover.ai/agents and aircover.ai/salesread 2026-09-01

Aircover documents both attestation and customer-facing controls on a dedicated trust center. SOC 2 Type II is stated as certified rather than in progress, alongside GDPR compliance.

Customer-facing controls: single sign-on for platform access, plus encryption of all data, authentication and tokens in transit and at rest, and configurable transcription notice options an account sets for its own regulatory position.

Around them sit per-customer data segmentation, an explicit statement that customer data is not sold or used to train third-party models, and a published privacy policy. The no-recording architecture is unusual and suits jurisdictions with strict recording rules.

Sourceaircover.ai/trust-centerread 2026-09-01

Aircover offers analytics about conversations, not a record of agent behavior. Conversation Intelligence, the Research Agent and the Deal and Coaching Intelligence dashboard surface trends, objections and win-loss patterns, and AI Coach scores calls against methodologies; transcripts and notes record what was said.

All of that describes the customer's meetings, which is the product's output. Nothing documents a run history, an agent activity log, a trail of which agent surfaced which play or wrote which CRM field, or any way to reconstruct why a given action was taken. That gap matters because the agents write to the CRM and send follow-ups.

Sourceaircover.ai/conversation-intelligence and aircover.ai/platformread 2026-09-01

Aircover retains deal and conversation history so Ask Aircover can answer about any deal in the pipeline and analysis spans all conversations. No dedicated agent memory is documented (aircover.ai sales).

Sourceaircover.ai/platformread 2026-10-01

Aircover is cloud only, with nothing the customer selects. No region choice, data center disclosure, VPC, private tenancy, self-hosted or on-premises option appears on the trust center, the platform pages or the integrations page; the downloadable desktop client that joins the virtual meeting is how the product reaches the call, not a deployment choice. Aircover operates without reliance on call recording, a choice that matters to buyers in regulated sectors.

Sourceaircover.ai/trust-centerread 2026-09-01

Aircover ships a named set of prebuilt specialist agents the customer adopts ready-made. It lists Live Coaching, Objection Handling, Virtual Sales Engineer, Call Prep, Procurement and a Research Agent, each with its own page, and frames them as enterprise-grade specialist agents deployed across the revenue org.

Alongside them sit three named methodology playbooks, MEDDPICC, SPICED and COVER, the latter two built with consultancy partners, which a team adopts rather than authors. Bespoke agents that Aircover builds with white-glove deployment alongside the customer's GTM engineering team are a services engagement rather than a packaged asset.

Sourceaircover.ai/agentsread 2026-09-01

Work reaches the agents by schedule and by event across the whole call lifecycle. Calendar sync with Google Calendar and Outlook drives pre-call preparation before a meeting exists in the product; the meeting itself is the live trigger, with full real-time support documented on Zoom, Microsoft Teams, Google Meet and Webex; and call end kicks off post-call actions automatically, syncing notes and CRM fields and generating follow-ups. Contact-center channels reach further through Twilio and Vonage, and dialer integrations with Salesloft and Nooks bring outbound calls into the same path.

Sourceaircover.ai/partners-integrations and aircover.ai/salesread 2026-09-01

Aircover uses generative AI in call, but the model is not disclosed and no customer model selection is documented (aircover.ai in-call-ai).

Sourceaircover.ai/platformread 2026-10-01

Aircover publishes an MCP server at api.aircover.ai/mcp with OAuth 2.0 (PKCE, dynamic client registration), a manifest and twelve enumerated tools, among them list_meetings, get_deal, ask_documents and agent_results, plus an OpenAPI file covering that endpoint. Every tool is read only, so nothing can create or change Aircover's own records through it, and no other API reference is published.

Sourceaircover.ai/developersread 2026-10-01

AI Coach scores rep calls against methodologies, which evaluates rep performance rather than testing or optimizing the agent itself (aircover.ai sales).

Sourceaircover.ai/platformread 2026-10-01

No browser or computer use capability is documented; Aircover assists live within meetings and syncs to the CRM.

Sourceaircover.ai/platformread 2026-10-01