habit-hooks-review
Spawn a reviewer sub-agent to assess a change set against habit-hooks's coding principles. Use AFTER habit-hooks reports clean —…
Write or revise a habit-hooks coaching prompt. Use when a linter / knip / jscpd rule fires and the agent's default fix is wrong or shallow, or when adding a project-local override prompt. Keeps prompts short and outcome-focused using the ROSE pattern.
$ npx -y skills add habit-hooks/habit-hooks --skill habit-hooks-prompting --agent claude-codeHow it fires
How this skill gets triggered: by you, by Claude, or both.
/habit-hooks-promptingContext preview
The summary Claude sees to decide when to auto-load this skill.
Write or revise a habit-hooks coaching prompt. Use when a linter / knip / jscpd rule fires and the agent's default fix is wrong or shallow, or when adding a project-local override prompt. Keeps prompts short and outcome-focused using the ROSE pattern.
name: habit-hooks-prompting description: "Write or revise a habit-hooks coaching prompt. Use when a linter / knip / jscpd rule fires and the agent's default fix is wrong or shallow, or when adding a project-local override prompt. Keeps prompts short and outcome-focused using the ROSE pattern."
Coaching prompts exist to **change behaviour the agent gets wrong by default** — nothing else.
The goal of every prompt is to drive the agent to **fix the root cause the violation points at, not to appease the reporting tool.** A fix that clears the report while going against the spirit of the rule is a failure, even when the tool goes green.
This is reactive: add a prompt the moment you catch the default behaviour being wrong, not pre-emptively for every rule.
A good prompt is the **minimum text that reliably nudges the right behaviour**. If it reads like an essay, cut it — direction beats explanation. The agent already has the violation in front of it.
ROSE structures feedback so the agent understands *why*, not just *what*, and fixes the cause rather than the symptom:
So a habit-hooks prompt writes **R, S, E** and lets the tool supply **O**. See `eslint-fatal.md` and `eslint-typescript-eslint-no-non-null-assertion.md` for prompts in this shape.
Turn best-practice coding advice into AI habits. Stop reciting software engineering literature to your AI agent.
Spawn a reviewer sub-agent to assess a change set against habit-hooks's coding principles. Use AFTER habit-hooks reports clean —…
Cut a new release of the habit-hooks packages. Use when asked to release, publish, or bump the version. Reviews what lands,…
Write or trim habit-hooks documentation under docs/. Use when creating, editing, reviewing or cutting any doc in this repo —…