Capability tokens
Task-scoped, short-lived (30-minute idle, 4-hour absolute by default), bound to tenant, user, mission and attempt, stored as HMAC hashes, revocable with cascade.
Security & trust
Built so agents can act without acting alone.
Insiro assumes every agent could be wrong and every tool could be misused, and designs controls for both. This page summarises them; our security team will walk through the details and your questionnaire.
Deployment
Run it where your data lives.
Fastest start
Shared
Multi-tenant SaaS with organisation-level isolation enforced in every query and every tool call.
Isolation
Dedicated
Single-tenant infrastructure provisioned for your organisation, with your own data stores.
Full control
Bring your own cloud
Deployed into your Azure subscription. Data, logs and secrets never leave your tenant.
Controls
Defense in depth for autonomous work.
Risk tiers and approvals
Tier 2 needs user confirmation; tier 3 needs owner approval and delegated credentials. Approval tokens are single-use and bound to the exact tool and arguments.
DLP
Input scanning can block sensitive data per tool. Outputs are always scanned and redacted before reaching a model or person.
Sandboxing
Code runs in isolated sandboxes with an egress allowlist, package allowlist, typosquatting detection and vulnerability scanning.
Safe cloud browser
Agent browsers cannot submit or post without approval, cannot reach private addresses, and never type into password or card fields.
Evidence and history
Append-only, hash-chained mission history and an evidence service that fails closed: if audit cannot be written, the call does not run.
Identity and access
Enterprise sign-in, roles, groups and Spaces; org admins manage members, models, tools and policies from one console.
Secrets
Pluggable secret backends including Azure Key Vault; data-source credentials encrypted at rest and masked in logs.
Controlled rollout
Feature flags with a policy engine let admins enable capabilities by org, group or percentage and roll back instantly.
Where does Insiro run and where is our data stored?
Insiro offers three deployment models: shared multi-tenant with strict organisation isolation, dedicated single-tenant infrastructure, and bring-your-own-cloud in your Azure subscription. With BYOC, data, logs and secrets stay in your tenant, and you can bring your own model keys so prompts go only to providers you already contract with.
How does Insiro handle credentials for connected apps?
Users connect their own accounts through OAuth. Tokens are stored encrypted in a pluggable secret backend such as Azure Key Vault, never exposed to models, and brokered by the gateway only at call time. Capability tokens for agents are stored as keyed hashes, so a database read never yields a usable token.
Which model providers see our prompts?
Only the providers your organisation enables. Admins choose which models are available, from Insiro-hosted models to OpenAI, Anthropic Claude and Google Gemini, and can bring their own API keys so prompts run under contracts you already hold. Organisation memory is stored per tenant and governed by your admins.
Bring your security questionnaire.
We'll walk through architecture, data flows and controls with your security team on a live workspace.
- 13governance checks on every agent tool call
- 1,700+SaaS apps agents can act in
- 30+data dialects for plain-English questions