Skip to content

Your team of AI agents, on your Mac, Windows or Linux. Agents that talk in channels, write code, review each other and use your tools, on the Claude Code, Codex, OpenCode or Pi you already have. Anything that touches the real world waits for your yes.

Let agents run on Gemini CLI (#25)

Eduardo Machadob2aa331267f0
diff --git a/src/core/harness.ts b/src/core/harness.ts
index 811ce50..d6aac90 100644
--- a/src/core/harness.ts
+++ b/src/core/harness.ts
@@ -7,7 +7,7 @@ import { L } from '../shared/lang'
import type { EngineId, EngineStatus } from '../shared/types'
// The harnesses the app can run agents on; one added later is listed as coming until then.
-export const RUNNABLE: EngineId[] = ['claude', 'codex', 'opencode', 'pi']
+export const RUNNABLE: EngineId[] = ['claude', 'codex', 'opencode', 'pi', 'gemini']
// Off for a harness pointed at a gateway: its own telemetry and error reports
// would go to Anthropic or OpenAI, around the gateway.
diff --git a/test/harness.test.ts b/test/harness.test.ts
index 8db1e96..ece7d67 100644
--- a/test/harness.test.ts
+++ b/test/harness.test.ts
@@ -22,6 +22,8 @@ describe('harnesses', () => {
expect(out.codex.loggedIn).toBe(true)
expect(out.opencode.loggedIn).toBe(true)
expect(out.pi.loggedIn).toBe(true)
+ expect(out.gemini).toMatchObject({ loggedIn: true })
+ expect(out.gemini.soon).toBeUndefined()
})
it('turns off the harnesses the person left out', () => {