02 · you ↔ your agents · what you do
Project Management
> Your intentions, executable.
A classic project management tool that doubles as the layer connecting a person to their agents — with the Data Base built in, so every task carries the knowledge needed to finish it.
# what it does
- -A classic project tool: goals, projects, cycles and tasks
- -The intermediate layer between you and your agents — assign work to either
- -Tasks linked to the knowledge they need in the Data Base
› Your workspace home: recent work, quicklinks and stickies.
⠋ loading fig/agent-dispatch…
[██░░░░░░░░░░░░░░]# connect it to your AI
endpoint: https://projects.avarile.com/api/mcp
sign-in: a personal access token, sent as a header
tools include: list_projects · create_work_item · update_work_item
# works with
# guides that use it
- Your weekly review, from the record
Five minutes a week: what moved, what got stuck, what you didn't plan, and three priorities for next week, built from what actually happened.
- Plan a goal into projects and tasks with an agent
Describe what you want; your AI breaks it into a project and tasks using how you planned the last one, links the know-how each task needs, and asks before it creates anything.
- Assign work to an agent and review it
Create an agent as a member of your project, assign it work items like a colleague, let it work through its queue, and approve the result before anything is done.
- Run cycles and get weekly reports drafted for you
Plan work in one- or two-week cycles, and let your AI draft the report from what actually happened in the cycle, then file it where you'll find it again.
- Finish a project properly: the check step and the retro
Nothing goes straight from doing to done. Add a check, run a short retro from the record, and make each lesson change a note, a template or a recipe.
- Link records across modules
Attach the right notes and records to a client in the CRM or a work item in Project Management, so the context travels with the work.