This page documents how TokenShift integrates with the model providers and services most commonly used alongside coding agents, and what admins can configure once connected.
Anthropic (Claude, Claude Code)
TokenShift sees Claude Code sessions on the workstation and reports token consumption and cost per developer, tied to repo, PR and ticket. Admins can allow or block which Claude models a team may use, and set budgets that warn, block or need approval.
OpenAI-based tools
The TokenShift integration with Codex is generally available. TokenShift reports Codex activity through the same admin console as its other supported agents: Claude Code, Cursor, GitHub Copilot CLI, OpenCode and Devin. The Devin and OpenCode integrations are also generally available.
Model routing policies
Smart Routing is a separate, opt-in module. It is not enabled by default.
Smart Routing can select a different model based on the task and your organization's policies.
Separately from Smart Routing, admins can allow or block models per organization, team or user, and set budgets that warn, block or need approval. This addresses a common governance question: keeping routine tasks from defaulting to the most expensive available model.
MCP server and tool governance
For teams running Model Context Protocol servers, TokenShift extends policy controls to individual tool calls: which MCP servers, tools and skills an agent is permitted to use on the workstation, set per team, with violations logged centrally. This gives InfoSec teams the governance and audit trail they typically ask for before broader rollout.
Frequently asked questions
How do we stop developers from defaulting to the most expensive frontier model when Claude Haiku or Sonnet would do the job?
Allow or block models per team, and set budgets that warn, block or need approval. Teams that want automatic model selection can opt in to the separate Smart Routing module and configure routing policies; when enabled, routing can select a model suited to the task within those policies.
We need policy controls on MCP servers and tool calls before our audit. What does this?
TokenShift's governance layer covers this directly: which MCP tools an agent may call, enforced per team, with every policy violation logged for audit review.
Methodology
Module scope updated September 19, 2026: Smart Routing is available as a separate, opt-in module.
Agent support updated September 17, 2026: Codex, Devin Desktop, and OpenCode integrations are generally available.
Naming updated September 23, 2026: Windsurf is now Devin Desktop.
Scope updated October 15, 2026: TokenShift covers AI agents on employee workstations. Supported agents are Claude Code, Codex, Cursor, GitHub Copilot CLI, OpenCode and Devin.
This guide reflects TokenShift's product documentation as of October 2026. For corrections, reach out at pointfive.co/contact.