architect
01Agent Claude Code
Plans changes to the skill pack — new sub-skills, addons, presets — respecting the ship boundary and the SKILL.md contract.
Turn any repository AI-first and run resumable, long-horizon Deep Work Plans with any coding agent — an installable, MIT-licensed agent skill.
This repository also configures its own agents. See what deepworkplan-skill tells them →
Agent Claude Code
Plans changes to the skill pack — new sub-skills, addons, presets — respecting the ship boundary and the SKILL.md contract.
Agent Claude Code
Implements a planned change to the skill pack task-by-task, running the local CI gate after each step.
Agent Claude Code
Audits and fixes SKILL.md / agent / command frontmatter against this repo's contract and the validator.
Agent Claude Code
Reviews a diff in this skill-pack repo for ship-boundary, frontmatter, bash 3.2, and public-surface violations before commit.
Agent Claude Code
Reviews and fixes the shipped shell scripts for shellcheck cleanliness and bash 3.2 (macOS) compatibility.
At most 3 mods per repository are shown here, and a mod shipped inside a plugin is left to that plugin's page — the rest are on their repository pages: