Launched this week
Kaiku
The task tracker your AI agents already know how to use
48 followers
The task tracker your AI agents already know how to use
48 followers
A task tracker and wiki built for teams whose work is increasingly done by AI agents. It speaks the API of the tracker most companies already run, so the MCP servers your agents use work unchanged. Built-in MCP server for Claude Code and Cursor, agents you call in a comment that propose but never decide, and every agent run's cost recorded on the issue it worked on.
Kaiku
Kaiku
The cost-on-the-issue part is what I'd want most. I run a bunch of scheduled agent jobs for a small job board (posting, moderation, a daily article), and the expensive surprises were never one big run, they were duplicates. My laptop slept through a few slots, the scheduler caught up on wake and the same job ran twice, so two identical posts went out and nobody noticed until a reader did. What fixed it was boring: each job reads a small log of what it already did today before touching anything. If an agent picks up the same Kaiku issue twice after a retry, does the second run see the first one's record and stop, or does it just log a second cost line next to it?