Skip to main content

Configure coding tools with Cortega AI Booster Pack

Use the extension in desktop VS Code on Windows or macOS to connect supported coding assistants to your Cortega gateway. Endpoint Guard on macOS adds Shadow AI controls when you need to monitor or block unapproved applications.

Set up​

  1. Install the Cortega AI Booster Pack extension in desktop VS Code and set cortega.ai.serverUrl to your Cortega server if your organization uses its own installation.
  2. Install the coding tools you want to use. Cortega detects Claude Code, Codex, Continue, OpenCode, Pi, Goose, Aider, and Cursor from installed editor extensions, existing configuration directories, or commands on PATH. Standalone tools need no VS Code plugin.
  3. Quit Cursor before configuring it. Open it once first if its configuration database does not yet exist. Keep Cursor closed when syncing or restoring its settings.
  4. Run Cortega AI Booster Pack: Sign in, complete your organization's sign-in, and review the tool terms shown by the extension. Cortega does not install standalone tools.
  5. Run Cortega AI Booster Pack: Sync AI tools with Cortega. In router mode Cortega uses the organization's default model alias. In direct mode, choose an available model first.
  6. Restart the configured clients or start fresh sessions. Reopen Cursor and select the configured model if needed. Check the Cortega status tooltip for individual tool errors and verify requests in gateway observability.

Coverage​

ToolWhat Cortega configures
Claude CodeShared CLI/extension gateway settings and legal MCP servers
CodexShared CLI/desktop/extension custom provider and legal MCP servers
ContinueLocal chat/edit/apply profile and legal MCP connection
OpenCodeCustom provider, authentication, default and small model
PiProvider registry, authentication, default provider/model
GooseDedicated custom provider and default provider/model
AiderOpenAI-compatible endpoint/key and main, weak, editor models
CursorSaved OpenAI BYOK base URL, key, and model selections

Continue autocomplete is not configured. The additional adapters configure model traffic; legal MCP provisioning remains limited to Claude Code, Codex, and Continue.

Cursor's adapter uses internal saved settings, which may change between versions. It refuses active SQLite journals or a running Cursor process. Cursor documents that BYOK requests still pass through Cursor's servers for prompt building, and Tab completion uses Cursor models. Configuring a private model does not make Cursor fully local or route every Cursor feature through Cortega. The gateway must be reachable through Cursor's request path.

Files and profiles​

The extension configures the local user's global profile. Project settings, environment variables, explicit CLI flags, separate profiles, and resumed sessions may override it. Those overrides are not prevented by the extension; gateway controls apply only to routed requests. Use Endpoint Guard on macOS for device-level Shadow AI controls.

  • OpenCode uses ~/.config/opencode/opencode.json (or existing opencode.jsonc) and ~/.local/share/opencode/auth.json; XDG config/data overrides are respected.
  • Pi uses ~/.pi/agent/{models,auth,settings}.json or PI_CODING_AGENT_DIR.
  • Goose uses its configuration directory and custom_providers/custom_cortega.json. On Windows that is %APPDATA%/Block/goose/config; macOS recognizes .config/goose and an existing Library/Application Support/Block/goose directory. GOOSE_PATH_ROOT redirects configuration to its config subdirectory.
  • Aider uses ~/.aider.conf.yml.
  • Cursor uses User/globalStorage/state.vscdb inside %APPDATA%/Cursor on Windows or ~/Library/Application Support/Cursor on macOS.

Repository inventory​

When an administrator ticks Collect repository inventory on Endpoint Guard → AI Assistants, Sync reports the Git repositories in this workspace and a bounded scan of your home directory: local path, remotes, organization, branch, Git identity, and working-tree state. Source code is not sent. Results appear on Endpoint Guard → Developer repositories. Sign in to one VS Code window is enough; the home scan includes checkouts that are open in other windows.

Undo and troubleshoot​

Run Cortega AI Booster Pack: Undo all changes to my AI tools to restore managed settings and pause management. Sign out also restores settings. Close Cursor first. If restoration fails, the extension reports the affected tools and retains recovery information so you can retry Undo.

New adapters preserve unrelated values and comments, record original managed values privately, and leave later user edits intact on undo. Credentials and recovery files are written with private permissions on Unix; Windows files inherit user-profile access rules.

Drift is checked every five minutes; settings are synced every thirty minutes. Run Sync to repair immediately. Malformed configuration is left unchanged and reported per tool, so one broken client does not prevent the others from configuring. For Cursor, run Sync from VS Code while Cursor is closed, then reopen it.

Projects and sensitive work​

Choose project selects a project from your organization's list; the gateway records the server-held selection on every request from your AI tools. The Cortega AI Booster desktop app uses the same list and the same selection.

A project can be marked sensitive by an administrator, or by whoever creates it. While you have a sensitive project selected, your requests are treated as privileged work. The team's privilege classifier can only choose models authorized as Sensitive, and a request with no suitable route is refused. You do not need to turn on the privilege switch yourself, and the Privilege/sensitive work command tells you the mark is always on until you choose another project. The status bar shows a lock while a sensitive project is selected.

At sign-in and sync, the extension registers the Projects MCP server (config name cortega-matters-projects) in Claude Code and Codex. Assistants can list, add, update/retire and bulk-import projects, including the sensitive flag. Reading and writing follow the caller's permissions; writes are audit-logged.

Enable the Project agent and provide active entries with useful descriptions (repository, product and task). After a quiet gap, the agent assigns matches with confidence of at least 80%; uncertain suggestions stay for review. Human session decisions take priority on reruns. The assignment does not change your selection for other work.

View Cortega Agents → Insights → Cost by project for gateway request counts, tokens and spend, plus usage telemetry estimates for Claude Code and Codex when their routing is off. The Project classifier need not be enabled. Explicit selections take priority; assigned gateway sessions follow, and unmatched usage appears as Unassigned. Classification calls are excluded. Choose a reporting window of up to 90 days.

In the VS Code AI Booster panel, open Project costs. The native AI Booster app shows the same report under Projects. If routing and usage telemetry are both off, the cost section is greyed out; you can still manage and select projects. Telemetry is allocated to the selected project when Cortega receives an export; delayed batches use that selection, and older unstamped exports are excluded. Codex costs are list-price estimates, not subscription invoices. Missing model prices are shown as unpriced.

See your usage, sessions and settings​

Three more commands open one panel inside VS Code. They show your own data from Cortega:

  • AI Usage: your Claude Code and Codex tokens and estimated cost, by model, for the last 7, 30 or 90 days.
  • AI Query Log: your sessions. Open one to read its prompts, responses and tool calls. Text appears only when your organization collects prompts and responses.
  • Settings: what Cortega has switched on: whether your AI tools are Routed through the Cortega AI Border Gateway, and whether Telemetry collects usage and prompts and responses. Your administrator sets these.

Open them from the Cortega status bar menu or the Command Palette.

On-prem tenant enrollment​

Use your app origin, not the gateway LLM URL, for cortega.ai.serverUrl. Sign-in selects an existing user's tenant first, then a claimed registration domain, and uses that tenant's IdP or the installation fallback. Register /api/v1/edge/sso/oauth/callback in the applicable OAuth client. Unknown unmatched users require approval; this sign-in flow does not create tenants.

After SSO the Pack registers VS Code's computer ID under Endpoint Guard → Seats, labelled VS. Repeated sign-in by the same person on the same computer reuses it; another person/computer uses another registration. The legacy ai_verifier_user role allocation and device-seat limits both apply. Native and VS registrations have distinct identities and can take separate seats. Deleting a row allows sync to recreate it; disable the account to prevent further registrations.

For environment variables, nginx host mapping, tenant capacity and acceptance checks, see On-prem multi-tenant setup.