- Product
- Ask Cyril
AI that runs across every module, not next to one.
Answers grounded in your own records. A plan you approve before Cyril updates deals, creates tasks and tickets, or raises invoices and chases payment. A model abstraction that never locks you to one vendor.
The foundation under every app
What Ask Cyril does for you
Cyril’s AI layer is not a product surface — it’s a layer that runs through every module. Answers from your own records, actions across Sales, Projects, Support and Finance that wait for your approval, and a model abstraction that’s never locked to one provider.
Generative answers, grounded
Core records — contacts, companies, deals, invoices, tickets, projects, docs and products — each have a view built for a model. Answers cite the records they read.
Ridgeline Property Trust. ROWA-0998 is overdue at 50,595.00, with their project due in five weeks.
Actions you approve
Ask Cyril updates deal stages, creates tasks and tickets, and raises and chases invoices, after you approve its plan. Each step is checked against your permissions before it runs.
Model abstraction
Anthropic Claude is the primary; OpenAI is the secondary. Add or swap providers in one place — callers never know which model answered.
Every run recorded
Every run is recorded: the plan, who approved it and what each step did. Nothing runs without your approval.
AI needs context. Cyril answers from your own records.
An assistant that can only see a CRM has thin context. Ask Cyril answers from your own records across Sales, Finance, Support, Projects and Docs, citing the records it read, and only those your role can open.
- Every step checked against your permissions before it runs
- A model-readable view of the core records, with more types being added
- An MCP connector for external agents is in development
Three modes, describing how hands‑on you choose to be.
Platform is the full application, driven by you — nothing happens that you did not do. AI is where you ask, and it plans, shows the plan, asks approval, and acts where you can watch. Agentic, where you hand a goal over entirely, is built but described as coming soon rather than generally available. The modes are about your involvement, not about how much AI is present: the layer is there in all three.
- Platform: you drive, nothing happens on its own
- AI: it plans, shows the plan, and acts with approval
- Agentic: coming soon, so don’t plan around it today
Common Ask Cyril questions
Can I bring my own LLM provider?
The model abstraction layer is designed for it. Adding a new provider is a single adapter file. At launch we ship Anthropic Claude as primary and OpenAI as secondary; dedicated self-hosted models (Sovereign AI) are in beta for Enterprise and set up per customer.
How do you prevent the AI from doing the wrong thing?
Ask Cyril shows its plan and waits for your approval. Each step is checked against your permissions on the server before it runs, and every run is recorded: the plan, who approved it and what each step did.
Can an agent see data the person asking cannot?
Not beyond their role. Ask Cyril only reads from modules the person can open, and every step of a plan is checked against their permissions on the server before it runs. That is also why a broad question returns less for a narrowly-permissioned user.
What happens when the AI gets something wrong?
You see the plan before anything runs, so most mistakes are caught at approval. Every run is recorded: the plan, who approved it and what each step did. Records it created or changed can be corrected like any other; a sent reminder cannot be recalled.
Be one of the first to use Cyril.
Join the waitlist for early access. We’ll only email you when there’s something real to share.
One email when your place opens. Nothing else.
