The Anthus Platform
Kanbus
Kanbus is a project-management system for orchestrating AI-agent workloads with git-backed issues, wiki-style planning, and durable task memory.
What it is
Kanbus is an issue tracker and planning system designed for a world where humans and AI agents collaborate directly in source-controlled workflows. Issues live as files in the repo, which keeps project memory portable, auditable, and easy for agents to read and update.
What it does
- Stores work items as plain files instead of locking them in a hosted task database.
- Exposes issue and planning workflows through CLIs and a local web console.
- Supports wiki-style project views that summarize live issue data.
- Uses vocabulary and structures that map cleanly to both human planning and LLM reasoning.
Why it matters
AI-agent orchestration needs durable work memory. If the tasks, notes, and status are trapped in a separate SaaS system, the engineering loop gets brittle. Kanbus keeps the work artifacts near the code and makes agent collaboration much easier to govern.
How it fits the Anthus Platform
Kanbus is the workload-orchestration layer for the broader platform. It complements Tactus by giving procedures a place in an actual delivery workflow, and it complements Primus by turning evaluation and platform work into a manageable operating cadence.
A common pattern is Kanbus + Tactus + Primus: track the work, run the procedure, and keep the quality loop visible over time.
Where this fits in the Anthus Platform
Kanbus supplies durable workflow memory so humans and agents can coordinate real delivery work instead of operating in a stateless loop.
- Pairs with Tactus for execution
- Feeds work cadence into Primus
- Appears in Korporus as an operator surface
Common combination
Example: combine Kanbus, Tactus, and Primus to manage backlogs, execute procedures, and continuously improve the resulting system.