---
title: "Claude Code · Blog · Projectify"
description: "Projectify articles on Claude Code."
url: https://projectify.si/blog/tag/claude-code
---

# Blog: Claude Code

Practical writing on running projects where Claude does real work.

- [Claude routines and scheduled tasks for a project board](https://projectify.si/blog/claude-routines-for-a-project-board.md) (2026-10-17): A Claude Code routine is a cloud session that keeps working while your laptop is closed. Wire it to the board and it wakes on two events, with a grace period, a cap and a record of every run.
- [Running several Claude Code agents without collisions](https://projectify.si/blog/running-several-claude-code-agents.md) (2026-10-17): Three sessions on one repo is where the afternoon goes. What a shared board changes: one worktree per claim, a 30-minute lease, a room, one queue, and a handoff at the end.
- [Why your AI agents need claims, not locks](https://projectify.si/blog/claims-not-locks.md) (2026-10-08): Two agents on one task is a lost afternoon. A lock fixes that and creates a new problem. A lease that frees itself fixes both.
- [Connecting Claude Code to your project board over MCP](https://projectify.si/blog/connect-claude-code-mcp.md) (2026-10-04): One command, one browser approval, and Claude can read your brief, pick up tasks and report back. Here is what happens at each step.
- [The handoff: end a Claude Code session so the next one starts briefed](https://projectify.si/blog/the-handoff.md) (2026-09-30): Most of what an agent learns in a session dies with it. A four-part handoff keeps it, takes the agent seconds to write, and is the first thing the next session reads.

Feed: https://projectify.si/rss.xml