# BOW

- [BOW](https://docs.bow.ai/introduction.md): Governed AI agents for any model and any data.
- [Quickstart](https://docs.bow.ai/quickstart.md)
- [Deployment Options](https://docs.bow.ai/install.md): Install BOW on your own server using Docker or Kubernetes
- [Day 2 Operations](https://docs.bow.ai/day-2-operations.md): Production guide: backups, high availability, upgrades and automatic updates, monitoring and security for running BOW
- [Sign-in and SSO](https://docs.bow.ai/sign-in-and-sso.md): Auth modes, OIDC and Entra ID, Google, SAML 2.0, and LDAP / Active Directory
- [Work with agents](https://docs.bow.ai/work-with-agents.md): Use governed agents for everyday questions, advanced investigations, recurring workflows, and shareable artifacts.
- [Chat with your data](https://docs.bow.ai/using-bow/reports.md): Prompt governed agents for questions, advanced analysis, root-cause investigation, and artifacts.
- [Dashboard artifacts](https://docs.bow.ai/using-bow/dashboards.md): Ask an agent to create, refine, share, and schedule a complete dashboard artifact.
- [Build apps with artifacts](https://docs.bow.ai/using-bow/artifact-apps.md): Create apps with saved records, private uploads and streamed AI responses through conversation.
- [Sub-agents](https://docs.bow.ai/using-bow/sub-agents.md): The agent delegates independent sub-questions to bounded background sub-agents that investigate on their own and report back as typed records in one turn.
- [Parameterized queries](https://docs.bow.ai/using-bow/parameterized-queries.md): Declare inputs on a query so a dashboard can filter it, and bind inputs to the viewer's identity for per-person data.
- [MCP](https://docs.bow.ai/using-bow/mcp.md): Connect Claude, Cursor, ChatGPT, or any MCP client to BOW with a personal key or an admin-registered OAuth app
- [BOW for Excel](https://docs.bow.ai/using-bow/excel.md): Ask BOW questions from inside Excel and drop AI-generated results straight into cells
- [AI Settings](https://docs.bow.ai/using-bow/settings.md): Configure AI capabilities and behavior for your organization
- [Agents](https://docs.bow.ai/agents/overview.md): Build governed domain agents from scoped connections, knowledge, tools, and access controls.
- [Instructions and knowledge](https://docs.bow.ai/agents/instructions.md): Manage reusable, governed agent knowledge across agents, data, channels, and modes.
- [Train an agent](https://docs.bow.ai/agents/train-an-agent.md): Teach an agent your data and business rules in a Training session, then review every change before it goes live.
- [Evals and self-improvement](https://docs.bow.ai/agents/evals.md): Verify agent behavior with suites and cases, then improve it through governed feedback loops.
- [Check-ins and overnight learning](https://docs.bow.ai/agents/self-improvement.md): Agents that follow up on their own and learn from repeated feedback overnight, within governed limits.
- [Lists](https://docs.bow.ai/agents/lists.md): Let an agent fill a typed table from documents, emails and data, with a source quote for every value.
- [Roles & permissions](https://docs.bow.ai/users/roles-and-permissions.md): Control who can do what with roles, resource grants, and groups.
- [Guides](https://docs.bow.ai/guides/overview.md): Step-by-step walkthroughs for common tasks in BOW.
- [Train an agent](https://docs.bow.ai/agents/train-an-agent.md): Teach an agent your data and business rules in a Training session, then review every change before it goes live.
- [Tips for training an agent](https://docs.bow.ai/guides/training-tips.md): What to teach an agent, how to write rules it follows, where each rule belongs, and how to keep it working as your data changes.
- [Create a demo dataset](https://docs.bow.ai/guides/demo-dataset.md): Ask the agent for realistic mock data for any domain, review the proposed schema and agents, and get a ready-to-use connection with agents on top.
- [Root cause analysis across observability tools](https://docs.bow.ai/guides/root-cause-analysis.md): Build one agent that searches Elasticsearch, Prometheus, Kubernetes and Splunk from an alert: train it on how your data fits together, give it a playbook, and get a root cause with evidence from every source.
- [Root cause analysis with sub-agents](https://docs.bow.ai/guides/root-cause-analysis-sub-agents.md): Ask why a metric dropped on the Checkout Ops demo and watch the agent test each candidate cause with a sub-agent, then name the cause and rule out the rest.
- [Build a dashboard with parameterized queries](https://docs.bow.ai/guides/parameterized-dashboard.md): Ask an agent for a dashboard with filters, add more filters in chat, and reuse the parameterized queries behind it.
- [Build a dashboard with state and saved data](https://docs.bow.ai/guides/stateful-dashboard.md): Build a Music Store album explorer where teammates leave ratings and comments that are saved, shared and summarized with AI.
- [Create a scheduled task for recurring analysis](https://docs.bow.ai/guides/scheduled-tasks.md): Run a prompt on a schedule, so a monthly digest or weekly check is ready without anyone asking for it.
- [Run an automated analysis from an external trigger](https://docs.bow.ai/guides/external-triggers.md): Give an alerting tool, CI job or script a webhook URL. Each event it sends starts a new session where the agent investigates and reports back.
- [Test an agent with evals](https://docs.bow.ai/guides/evals.md): Write test questions with expected answers, run them against an agent, and read why each one passed or failed.
- [Analyze your agents' usage in chat](https://docs.bow.ai/guides/usage-analysis.md): Ask a training session about your agents' run history — runs, failures, feedback, cost — and get charts, tables and instruction fixes back.
- [Use saved queries for faster, more reliable answers](https://docs.bow.ai/guides/saved-queries.md): Save a reviewed query once, and the agent loads it instead of writing new SQL every time someone asks the same question.
- [Route questions to the right agent](https://docs.bow.ai/guides/agent-routing.md): Help BOW pick the right agent when a question could fit more than one, with clear agent descriptions and one global routing instruction.
- [Analyze documents with Lists](https://docs.bow.ai/guides/lists-documents.md): Point an agent at a folder of contracts, have it fill a typed list with a source quote for every value, review and fix the rows, re-run after a change, then chart and export the results.
- [Use BOW from Claude or another MCP client](https://docs.bow.ai/guides/mcp-clients.md): Connect Claude Code, Claude Desktop, Cursor, or Claude on the web to BOW and ask questions about your data from your assistant.
- [Git Workflow](https://docs.bow.ai/guides/git.md): Manage instructions as code with Git integration and CI/CD automation
- [External Entra applications](https://docs.bow.ai/guides/external-entra-applications.md): Sign users into your application and access BOW and Power BI with delegated Entra identity
- [Completions API](https://docs.bow.ai/guides/completions-api.md): Create a conversation, ask questions, and choose the model, reasoning effort and tools it uses
- [Embed a chatbot in your ERP or portal](https://docs.bow.ai/guides/embed-chatbot.md): Build a branded BOW chatbot, configure Connected applications and authentication, and add a floating assistant to your ERP or portal with the embed SDK.
- [Data Sources Overview](https://docs.bow.ai/data-sources/overview.md): Every connector BOW supports, and the three kinds of connections you can make
- [Connecting a Data Source](https://docs.bow.ai/data-sources/connecting.md): The setup flow: connect, choose tables, add context, and control access
- [Authentication and Access](https://docs.bow.ai/data-sources/authentication.md): Shared credentials vs per-user sign-in, OAuth, Kerberos SSO, and what requires Enterprise
- [SQL databases](https://docs.bow.ai/data-sources/connectors/databases.md): Reference for the database and query-engine connectors that need only network access and a login
- [Snowflake](https://docs.bow.ai/data-sources/connectors/snowflake.md): Connect a Snowflake warehouse so the agent can explore its schemas and run SQL
- [Google BigQuery](https://docs.bow.ai/data-sources/connectors/bigquery.md): Connect BigQuery datasets so the agent can read their schemas and run SQL, with per-query cost caps
- [Microsoft SQL Server](https://docs.bow.ai/data-sources/connectors/sql-server.md): Connect on-prem or cloud SQL Server, with a SQL login, service-account Kerberos, or per-user Kerberos SSO
- [SAP](https://docs.bow.ai/data-sources/connectors/sap.md): Connect SAP HANA, SAP BW, SAP Datasphere, and SAP BusinessObjects to BOW
- [Power BI](https://docs.bow.ai/data-sources/connectors/power-bi.md): Connect Power BI semantic models to BOW and query them with DAX
- [Power BI Report Server](https://docs.bow.ai/data-sources/connectors/power-bi-report-server.md): Connect an on-premises Power BI Report Server and query its reports, datasets, and PBIX models
- [Microsoft Fabric](https://docs.bow.ai/data-sources/connectors/microsoft-fabric.md): Connect Microsoft Fabric Warehouse and Lakehouse SQL endpoints and query them with T-SQL
- [Microsoft Analysis Services](https://docs.bow.ai/data-sources/connectors/analysis-services.md): Connect SSAS Multidimensional and Tabular models and query them with MDX or DAX over XMLA
- [Tableau](https://docs.bow.ai/data-sources/connectors/tableau.md): Connect Tableau published data sources and let the agent query them through VizQL
- [Qlik Sense](https://docs.bow.ai/data-sources/connectors/qlik-sense.md): Connect a Qlik Cloud tenant and let the agent query its apps with hypercubes
- [Qlik Sense on-prem](https://docs.bow.ai/data-sources/connectors/qlik-sense-onprem.md): Connect Qlik Sense Enterprise on Windows with a QMC certificate and let the agent query its apps with hypercubes
- [Sisense](https://docs.bow.ai/data-sources/connectors/sisense.md): Connect Sisense data models and query ElastiCubes with SQL
- [Oracle BI](https://docs.bow.ai/data-sources/connectors/oracle-bi.md): Connect Oracle BI subject areas and query them with Logical SQL
- [Infor OLAP](https://docs.bow.ai/data-sources/connectors/infor-olap.md): Connect Infor d/EPM OLAP cubes over the XMLA Provider and query them with MDX
- [Timbr AI](https://docs.bow.ai/data-sources/connectors/timbr.md): Connect a Timbr ontology and query concepts, relationships, and measures with SQL
- [Observability and monitoring](https://docs.bow.ai/data-sources/connectors/observability.md): Connect Splunk, Elasticsearch, OpenSearch, Jaeger, Zabbix, Prometheus, AppDynamics, AWS CloudWatch, AWS Cost Explorer and Azure Cost Management so the agent can investigate logs, traces, metrics and spend
- [Kubernetes](https://docs.bow.ai/data-sources/connectors/kubernetes.md): Investigate a cluster read-only — workloads, nodes, events, networking, storage, custom resources, live usage and pod logs as tables
- [Microsoft Intune](https://docs.bow.ai/data-sources/connectors/microsoft-intune.md): Query Intune device management read-only — managed devices, compliance, configuration profiles, apps, discovered software and Autopilot
- [Microsoft Configuration Manager (SCCM)](https://docs.bow.ai/data-sources/connectors/configuration-manager.md): Query SCCM / MECM read-only — devices, collections, hardware and software inventory, applications, deployments and software update compliance
- [CrowdStrike Falcon](https://docs.bow.ai/data-sources/connectors/crowdstrike.md): Investigate endpoint security read-only — hosts, alerts, incidents, CrowdScore, Spotlight vulnerabilities, prevention policies and custom IOCs
- [NetApp ONTAP](https://docs.bow.ai/data-sources/connectors/netapp-ontap.md): Investigate storage incidents with read-only ONTAP inventory, topology, performance, events and diagnostic tables
- [Brocade Fabric OS](https://docs.bow.ai/data-sources/connectors/brocade.md): Investigate Fibre Channel ports, optics, zoning, congestion and events on a SAN switch with read-only diagnostics
- [VMware Aria Operations](https://docs.bow.ai/data-sources/connectors/aria-operations.md): Query the vSphere estate and every installed management pack — inventory, topology, metrics with dynamic thresholds, alerts and symptoms
- [Business applications](https://docs.bow.ai/data-sources/connectors/business-apps.md): Connect Salesforce, NetSuite, ServiceNow, Monday, PostHog and Priority ERP so the agent can query them as tables
- [Files and Directories](https://docs.bow.ai/data-sources/connectors/files-and-directories.md): Expose a local folder or a mounted network share (SMB/NFS) to the agent as a searchable file catalog
- [Amazon S3](https://docs.bow.ai/data-sources/connectors/s3.md): Expose an S3 bucket (or a prefix within it) to the agent as a searchable file catalog
- [SharePoint and OneDrive](https://docs.bow.ai/data-sources/connectors/sharepoint-onedrive.md): Read files from a SharePoint document library or from each user's own OneDrive via Microsoft Graph
- [SharePoint Server (on-prem)](https://docs.bow.ai/data-sources/connectors/sharepoint-server.md): Read and search on-premises SharePoint Server document libraries over the REST API with NTLM or Kerberos — no Microsoft Graph or Entra app required
- [OpenText Documentum](https://docs.bow.ai/data-sources/connectors/documentum.md): Browse, search and read documents from a Documentum repository over REST Services, with the repository's own ACLs applied to every read
- [OneNote](https://docs.bow.ai/data-sources/connectors/onenote.md): Search and read Microsoft OneNote notebooks — pages become a searchable catalog the agent can grep and read, embedded images included
- [SharePoint Lists](https://docs.bow.ai/data-sources/connectors/sharepoint-lists.md): Query SharePoint lists as tables — every list on a site becomes a queryable table with typed columns and OData filters
- [Google Drive](https://docs.bow.ai/data-sources/connectors/google-drive.md): Let each user connect their own Google Drive so the agent can read their Sheets, Excel, CSV and documents
- [Gmail and Outlook Mail](https://docs.bow.ai/data-sources/connectors/mail.md): Let the agent search and read a user's own mailbox as a data source
- [CSV, QVD and PBIX](https://docs.bow.ai/data-sources/connectors/file-formats.md): Turn CSV, Qlik QVD and Power BI .pbix files on the backend host into SQL tables the agent can query
- [Browser](https://docs.bow.ai/data-sources/connectors/browser.md): Let an agent read and interact with a specific set of web pages in a real headless browser, scoped to a URL allowlist
- [MCP Connections](https://docs.bow.ai/data-sources/mcp.md): Connect any MCP server, then govern which tools the agent may call and when
- [Custom API Connections](https://docs.bow.ai/data-sources/custom-api.md): Turn any REST API into agent tools by describing its endpoints
- [User Context Forwarding](https://docs.bow.ai/data-sources/user-context.md): Pass the caller's identity to an MCP server on every tool call, without the model being able to change it
- [Dynamic Params](https://docs.bow.ai/data-sources/dynamic-params.md): Send per-question values from your app to an MCP server on every tool call
- [Overview](https://docs.bow.ai/bow-fast/overview.md): Prepare a database for agent analysis: a governed, materialized copy that is fast to query and filtered per person
- [Custom queries](https://docs.bow.ai/bow-fast/custom-queries.md): Write the SQL once, put it on a schedule, and choose which agents may use it
- [Row-level security](https://docs.bow.ai/bow-fast/row-level-security.md): Filter a shared materialized copy per person, against their profile attributes, groups, or roles
- [Cut ClickHouse Cloud costs](https://docs.bow.ai/bow-fast/clickhouse-cloud-costs.md): Let agents answer from cached custom tables so an idle ClickHouse Cloud service stays idle
- [Models and LLM providers](https://docs.bow.ai/using-bow/llm.md): Bring your preferred models, route work intentionally, and keep a reliable fallback path.
- [Observe and govern](https://docs.bow.ai/observe-and-govern/overview.md): Trace agent work, understand quality and cost, and operate BOW with confidence.
- [Runs, traces, and diagnostics](https://docs.bow.ai/observe-and-govern/runs-and-traces.md): Inspect the evidence and decisions behind an agent outcome.
- [Usage, cost, and quality](https://docs.bow.ai/observe-and-govern/usage-cost-quality.md): Measure adoption, latency, spend, and reliability across your agent fleet.
- [Quotas and usage limits](https://docs.bow.ai/observe-and-govern/quotas-and-usage-limits.md): Cap monthly LLM tokens, spend, queries, and data volume per user, group, or role.
- [Audit logs](https://docs.bow.ai/observe-and-govern/audit-logs.md): See who did what in your workspace, export it, and stream every event to your SIEM or storage bucket.
- [Channels Overview](https://docs.bow.ai/channels/overview.md): Bring the BOW analyst to messaging tools and connect your own applications with OAuth
- [Slack](https://docs.bow.ai/channels/slack.md): Let users ask data questions via Slack DMs and channel mentions — no public URL required with Socket Mode
- [Microsoft Teams](https://docs.bow.ai/channels/teams.md): Let users ask data questions directly via Teams DMs and channel mentions
- [Google Chat](https://docs.bow.ai/channels/google-chat.md): Chat with the analyst from Google Chat — outbound-only Pub/Sub connection, no public URL required
- [WhatsApp](https://docs.bow.ai/channels/whatsapp.md): Let users ask data questions over WhatsApp via the Meta WhatsApp Business Cloud API
- [Email](https://docs.bow.ai/channels/email.md): Give the analyst its own mailbox — send answers by email and (optionally) receive questions
- [OAuth Apps](https://docs.bow.ai/channels/oauth-apps.md): Let your own applications sign users in to BOW and call the BOW API with OAuth 2.1 and PKCE
- [Open source and Enterprise](https://docs.bow.ai/enterprise/overview.md): Compare the self-hosted open-source platform with Enterprise governance and scale controls.
- [License Key](https://docs.bow.ai/enterprise/license.md): Activate Enterprise Edition by configuring your license key


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.