#Automation

Codex on Astra hit the weekly usage cap in under 12 hours. Fixes: 120s minimum wait in multi_agent_v2, an 8-field JSON receipt between agents, Astra demoted to advisor, Sol high on top.

Backlog issue statuses double as AI-agent states: auto-triage, local-only review, parallel git worktrees, and serial patch merge. Three steps stay manual.

A breakdown of Claude Code Routines, released as a research preview on April 14, 2026. Covers the three trigger types (schedule, API, GitHub), routine structure, and autonomous execution on Anthropic-managed infrastructure.

Generalized the scripts from the practice and optimization articles into a reusable framework and published it on GitHub. A walkthrough of how to use it and the design philosophy.

Design patterns for reducing context usage and API calls in the AI auto-dev loop: blocking waits, read-forbidden files, and session isolation.

Running the Claude Code + Codex auto-loop for real. Generated 1,134 lines of game code.

Technical prep for automating an implement → review → fix loop with Claude Code and OpenAI Codex via tmux. Can it build something overnight unattended?

How to call Xcode CLI tools from Claude Code to automate iOS builds, device installs, and App Store uploads.

How to set up an environment using the ralph-wiggum plugin to automatically run tests overnight and just review the results in the morning.