Questions, answered directly.

Insiro is an AI operations workspace for growing businesses. It brings an assistant, digital teammates, no-code agents, workflows, business-data analysis, BI reporting, knowledge, collaboration and controls into one place, so a lean team can move from a request to an accepted, evidenced result.

The clearest fit is a growing team with repeatable digital work, connected business tools and an owner who wants measurable capacity, not another experiment. If recurring work crosses email, documents, a CRM and a database before anyone can call it done, Insiro is built for you.

Digitally active firms with roughly 20 to 250 employees are the strongest fit, especially professional services, B2B SaaS, finance and accounting, ecommerce, wholesale, distribution and multi-location services. They have repeatable work and several apps, but no large internal AI or automation team.

No. Most work starts in plain language: chat, define a goal, generate a pipeline, teach a browser skill, ask a data question or describe a report. Developer and custom-tool options are there for advanced needs, but a typical team gets its first workflow running without writing code.

Yes. Insiro supports app permissions, teammate capability ceilings, human approvals, review queues, budget and credit limits, run history, evidence and operational telemetry. Routine read-only work keeps moving; consequential actions pause for the right person with the evidence attached.

Insiro Platform is five products on one harness, one company-context layer and one governance layer: Agent Harness for reliable execution, Datasight for data and BI, Multiplayer for coordinated AI teammates, Automation for cross-app workflows, and Governance for MCP controls and telemetry.

Insiro Automation uses a managed catalog of more than 1,700 applications alongside native MCP connectors, and Datasight supports more than 30 database and warehouse dialects plus MongoDB. Availability depends on enabled connectors, credentials and workspace configuration.

Insiro agents work across more than 1,700 business apps, including Salesforce, HubSpot, Gmail, Outlook, Slack, Microsoft Teams, Jira, Zendesk, ServiceNow, Notion, Google Drive, NetSuite, QuickBooks, Stripe and Workday, plus any MCP server you register. The most common journeys pair email with the CRM, support with issue tracking, and billing with the ledger.

Insiro Datasight answers plain-English questions over SQL Server, Oracle, PostgreSQL, MySQL, TiDB, Azure SQL, Amazon Redshift, IBM Db2, Kusto, Microsoft Fabric, Power BI, Teradata, Snowflake, Databricks, BigQuery, Cassandra, DynamoDB, Cosmos DB and Elastic, as well as Excel workbooks and PDFs. Queries are read-only and every answer keeps its SQL and source trace.

Choose one recurring process with a clear owner and a measurable result, such as a daily operations digest, lead research and follow-up, invoice exception review, support triage or management reporting. Run it as a pilot, measure the capacity returned, then reuse what works.

An agent harness is the runtime around a model that turns it into a dependable worker: it supplies tools, memory and knowledge, routes each request, enforces permissions and budgets, verifies results and records what happened. Insiro Agent Harness provides all of that so teams build agents instead of infrastructure.

Insiro routes across Insiro-hosted models and models from OpenAI, Anthropic Claude and Google Gemini. Organisations can bring their own API keys (BYOK), set a default model per agent or teammate, and let the router pick a fast, reasoning or coding model per request based on intent and cost.

Yes. Agent Harness imports agents from OpenAI Assistants, Claude and Gemini configurations or a JSON definition, keeping their instructions and mapping their tools to Insiro connectors. Imported agents immediately run under the same governed gateway, budgets and audit trail as native agents.

Every planned action is classified into four risk tiers. Tier 0 and 1 run immediately; tier 2 needs user confirmation; tier 3 needs owner approval and the user's own credentials. Plans containing risky steps are rejected unless they include an approval gate, and each agent carries a task-scoped capability token that limits tools and arguments.

Skills are reusable procedures taught by demonstration. You perform a task once in Insiro's cloud browser, Insiro records the steps, apps and guardrails, and agents can then run that skill on request or automatically when relevant. Skills can be shared with people, groups or the whole organisation.

Yes. Agents run code in an isolated sandbox with an egress allowlist, package policy and vulnerability scanning, and a coding agent gets a full development workspace per run. Results come back as files, tables or answers; destructive commands such as rm -rf or terraform destroy are classed tier 3 and need approval.

Insiro Multiplayer lets you work alongside AI teammates. You describe an outcome, set a budget and choose which apps may be used. A lead teammate drafts a plan, builds a small team of specialised sub-agents, does the work in those apps and asks you only when it needs a decision or input.

A goal is the outcome you want, with acceptance criteria and a spending cap. A teammate is a persistent AI colleague with a role, persona, model and an app ceiling. For each goal, the lead teammate creates sub-agents, such as a researcher, analyst or writer, that can never use apps beyond the teammate's ceiling.

Yes. Start from role templates such as Sales Researcher, Talent Scout, Paid Media Analyst, Chief of Staff, Support Triage or Engineering Debugger, or define your own. Each teammate has a model, a default budget and the apps it may use. You can pause, edit, archive or give a teammate a new goal at any time.

You approve the plan before work starts, see every task and spend live, and can steer the lead teammate with a message at any time. Apps you grant are used without asking again for that goal; anything held for a decision lands in your Needs you queue, and approving runs exactly what you saw, once.

You write acceptance criteria when you create the goal, one per line. Every task's output is verified, and the final result is checked against each criterion and marked met or not met with a reason. If steps fail, Try again re-runs only what failed and keeps finished work.

Yes. Any completed or partly completed goal can be turned into a routine with one click, so the same team runs it on a schedule, for example a weekly competitor digest or a monthly close checklist, with the same budget, apps and approvals.

Insiro Governance is the control plane for AI agents and MCP tools. Admins set each tool to off, read-only or read and write with clicks, require approvals by risk tier, apply kill switches and DLP, and use telemetry to see which tools ran for which user query and how many tokens each step consumed.

MCP governance means controlling which Model Context Protocol tools AI agents can discover and call, with what arguments, under whose credentials, and recording every decision. Insiro routes every MCP call through one gateway that enforces permissions, rate limits, risk-tier approvals, DLP and budgets, and writes audit evidence.

Open MCP controls in the admin console, find the tool or server, and choose Read only. The policy is pushed to the gateway immediately: write actions from that tool are blocked for every agent, teammate and pipeline in scope, and the change is recorded in the audit trail with who made it.

Telemetry shows active agents, success rates, latency, error rates, credits and token usage over time, plus a trace for every run. Each trace lists the user's query, the routed intent, every tool call with its permission decision, and input and output tokens and cost at each step.

Yes. Org admins have per-tool kill switches in the governance console, and platform kill switches can target a tool, a user, a workspace or a whole organisation. Kill switches take effect on the next tool call; the platform switch fails closed if its state cannot be read.

Yes. Inputs to tools are scanned and can be blocked per tool policy. Outputs are always scanned and redacted, covering identifiers such as social security numbers, card numbers, email addresses, password or token assignments and long secrets, before results reach the model or the user.

Every goal and mission has one budget account with an approved ceiling. Model calls, tools, sandbox time, search and storage reserve estimated cost before running and settle actual usage afterwards. When the ceiling is reached, work stops cleanly and the itemised spend is visible in telemetry.

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.

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.

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.