Plan: integrations are in
1 file+6−10/1 viewed
| 771 | 771 | and comments stand out in the feed; agents hold g1t's tools through a token | |
| 772 | 772 | scoped to one repository), steering a running agent (messages delivered | |
| 773 | 773 | between steps, and at the end), and recording sessions from anyone's own | |
| 774 | − | Claude Code (`curl -fsSL https://g1t.sh/install/claude.sh | sh`). Racing a | |
| 774 | + | Claude Code (`curl -fsSL https://g1t.sh/install/claude.sh | sh`). Integrations are | |
| 775 | + | in: a workspace's own model provider (Anthropic or any Anthropic-compatible | |
| 776 | + | endpoint, reached through a model proxy so no sandbox holds a key, for a | |
| 777 | + | flat orchestration fee), alerts from Sentry, Datadog and signed webhooks | |
| 778 | + | that open one issue per problem and can start an agent, and Jira and Linear | |
| 779 | + | tickets that agents read, people import, and that hear back. Racing a | |
| 775 | 780 | set number of agents on one issue is dropped: choosing how many agents to | |
| 776 | 781 | use is not something people should have to do. | |
| 777 | 782 |