A maintainer agent that reads open issues, searches the repo for related code, and proposes labels and assignees. Closing or commenting are destructive operations, so they pause for operator approval when HITL is enabled — useful for onboarding maintainers without giving them write access on day one.
Score file
import { defineScore, AnthropicProvider } from '@tuttiai/core'
import { GitHubVoice } from '@tuttiai/github'
export default defineScore({
provider: new AnthropicProvider(),
agents: {
triage: {
name: 'triage',
model: 'claude-sonnet-4-6',
system_prompt:
'You triage open GitHub issues. Search the repo for context, propose labels and assignees, and write a short summary comment. Do NOT close issues without explicit operator approval.',
voices: [new GitHubVoice()],
permissions: ['network'],
},
},
}) How to run it
- 01 Scaffold a project:
npx tutti-ai init my-github-triage - 02 Install voices:
npm i @tuttiai/github - 03 Drop the score above into
tutti.score.ts, set the env vars in.env, and runtutti-ai run.
Why this is safe
-
HITL on labels, comments, and closes -
permissions: ["network"]