Running Opair as a service, AI Code Tools

Opair Agency Implementation, Delivering Reviewed AI Code to Clients

Learn how to set up Opair's driver/navigator model in client projects, configure role-based permissions via JSON, and build a repeatable delivery workflow that keeps humans in control of every code change. This course teaches agencies how to productize AI-assisted development while maintaining code quality and client trust through enforced review gates.

Open the decision record for Opair

What does running Opair for clients commit you to?

Published figures for this service. Blank fields are not published.

Monthly tool cost
Opair is open-source; vendor software cost is not published. The lowest paid plan is not applicable. Agency investment is in setup labor and LLM provider usage costs, which are not modeled from supplied data.
Time to first value
Not published
Payback
Not modeled (client price, labor, usage cost, overhead, and expected volume not supplied)
Guided implementation
8 hours

Is Opair worth running as a client service?

Opair supports a managed oversight service for agencies prioritizing code quality and human review over maximum automation. Vendor pricing and time-to-value are not published, so agency economics must be measured through actual delivery hours, LLM usage costs, and client-approved scope.

An agency-fit judgement for reselling this service. It is separate from the tool description on the decision record.

Before you start

What has to be in place before the first client engagement.

Tools and subscriptions

  • Opair harness installed on a development environment
  • LLM provider credentials (OpenAI, Anthropic, or local models via llama.cpp)
  • GitHub, GitLab, or other MCP-compatible repository access
  • Slack, Linear, or Atlassian workspace for MCP integrations

People and inputs

  • Role configuration JSON files for driver, navigator, plan, document, and review roles
  • Tool definition JSON files for custom engineering tools
  • Editable prompt text files for hot-reloading during sessions
  • Technical staff to configure restricted tool scope and approval gates

Included with the course

7 working documents for delivering this service.

  • Opair Role Configuration Template for Client Projectstemplate
  • File-Write Approval Policy Checklist by Risk Levelchecklist
  • Driver/Navigator Workflow SOP for Agency Teamssop
  • LLM Provider Setup and Credential Management Worksheetworksheet
  • Custom Tool Definition Guide for Engineering-Only Scopeguide
  • Client Onboarding Handoff Document for GitHub/GitLab Integrationtemplate
  • Prompt Hot-Reload Change Log and Rollback Proceduresop

Listed by name. These documents are not yet published as individual downloads.