AI chat
Ask questions with project, docs, ticket, and memory context.
AI and agents
Give AI useful context without giving it the final word.
TOW combines permission-aware retrieval, configurable agents, and reviewable proposals across projects, docs, and memory.
How it works
Context in, review out.
Keep the AI path inspectable from source to accepted change.
- 01
Ground
Retrieve accessible workspace context and resolve references.
- 02
Assist
Answer, draft, summarize, detect conflicts, or run configured agent work.
- 03
Review
Route consequential proposals through a review flow unless an administrator has explicitly enabled a narrow direct-write agent path.
Capabilities
AI across the workspace.
Choose the jobs, models, and boundaries that fit the organisation.
Configurable agents
Define agent roles and run repeat operating workflows.
Provider controls
Configure organisation AI, BYOK endpoints, or disable AI.
Proposal inbox
Inspect proposed docs, work, and memory changes before acceptance.
Control point
Provider control stays with administrators.
Organisations choose the enabled provider configuration and remain responsible for reviewing provider terms, endpoint configuration, contracts, and the data sent by enabled workflows.
- BYOK options
- Configurable endpoints
- Organisation controls
- Full AI disable switch
Deployment and security
Control the deployment and the boundary.
Use TOW Cloud, run TOW on controlled infrastructure, or operate an air-gapped Docker Compose deployment. Permissions, identity settings, exports, and AI provider choices stay visible to administrators.
Explore self-hosting
FAQ
Questions worth answering clearly.
Current product, deployment, and commercial boundaries.
Can administrators disable AI?
Yes. TOW includes server, organisation, and user-facing AI controls, including a full disable path.
Does TOW support bring your own key?
Yes. BYOK plans let organisations configure their own compatible AI provider credentials and endpoints.
See TOW in your environment