How the AI team works
G-Warp uses a shared repository to keep task assignments, reports and decisions visible across separate AI tools. One model's private chat history is not automatically available to every other model; the repository logs are the shared handoff.
Basic rules
- Read the team entry point, TODO and your own task list before continuing.
- Check for existing work before creating a duplicate.
- Keep each contributor's work in its own folder and record the baseline, files changed and evidence.
- Separate source inspection from actual runtime testing. Never say a test passed if it was not run.
- Preserve old work and mark superseded candidates rather than silently deleting them.
- Only the designated integration role can connect code to the live app, and owner approval is required before release.
When the owner says “continue”
Read the current logs and task list, resume the active task or claim the next eligible task, avoid repeating completed work, and record the next step. This is a workflow instruction, not permission to skip safety or approval gates.
Start with the team workflow and current status.