Skip to main content
skill CX Codex

05-rule-generate — Claude Code skill: coding rules for AI hosts

05-rule-generate takes a coding rule from draft to the rule files of installed AI hosts: create, edit, publish, delete.

MIT ★ 571 updated 2026-10-29

Install into Claude Code · Plugin marketplace All methods and platforms ↓
/plugin marketplace add ai-driven-dev/framework
/plugin install aidd-context@aidd-framework
Open on GitHub

What 05-rule-generate does

Two actions do the work. capture-rule frames the request, compares it against what the project already holds, and prepares a candidate; write-rule writes the approved version and verifies it. Nothing lands on disk until you approve the candidate shown last. A pick, a merge, an edit or a rename does not count as approval — the skill displays the revision and waits again.

The argument is topic | auto, so either you name the topic or the skill picks one. Hosts it does not support come back as a report with alternatives, and a single AGENTS file tells it no host is installed. An installed reference it cannot read halts the run and gets named.

Existing sources and publications change only with explicit permission that names them. Keep or cancel leaves project files as they were. Skills, subagents and hooks fall outside its scope.

What it helps with

  • Add a naming convention to a project
  • Revise an existing style rule and roll it out to each host
  • Drop an outdated rule without touching neighbouring files
  • Find out which AI hosts are actually installed
  • Check that the stored rule matches the approved candidate

Example prompts for the agent

  • Add a rule that sorts imports alphabetically, topic imports.
  • Check which hosts are installed and write the rule for each, argument auto.
  • Delete the logging rule, but show me the candidate first.

How to install 05-rule-generate

1Plugin marketplace

/plugin marketplace add ai-driven-dev/framework
/plugin install aidd-context@aidd-framework

This installs the whole aidd-context plugin with all its skills

2Skills CLI (Vercel)

npx skills add ai-driven-dev/framework --skill 05-rule-generate -a claude-code

3Manually: put the files in place

~/.claude/skills/05-rule-generate/SKILL.md

Copy the whole skill folder; for one project use .claude/skills/05-rule-generate/

Step by step, with commands for each platform: How to install skills in Claude Code: five ways

Questions about 05-rule-generate

How do I install 05-rule-generate in Claude Code?

Through the marketplace: run /plugin marketplace add ai-driven-dev/framework, then /plugin install aidd-context@aidd-framework. The other route is npx skills add ai-driven-dev/framework --skill 05-rule-generate -a claude-code.

What does 05-rule-generate do?

It creates, updates, publishes or deletes a coding rule across the AI hosts installed in the project. It prepares a candidate for you to review, then writes the approved version and verifies the result.

Does it write changes without asking?

No. It writes only once you approve the candidate shown last, and an edit or rename in your reply counts as a revision rather than approval. Existing sources and publications need explicit permission that names them.

Source text

SKILL.md · MIT
---
name: 05-rule-generate
description: Create, update, publish or delete a coding rule across host AI tools. Use when adding, changing or removing a rule, convention or coding standard. Not for skills, agents or hooks.
argument-hint: topic | auto
---

# Rule Generate

```mermaid
flowchart TD
  request[Rule request] --> capture-rule
  capture-rule -->|create/update| decision[Assessed candidate]
  decision -->|revise/clarify| capture-rule
  decision -->|keep/cancel| stop
  decision -->|approved| write-rule
  capture-rule -->|publish/delete confirmed| write-rule
  write-rule -->|revised| capture-rule
  write-rule -->|refused| stop
  write-rule -->|verified| result
```

## Actions

Follow the flow. Read only the next action and its references.

| Action | Does |
| --- | --- |
| capture-rule | frame, compare, plan, decide |
| write-rule | write the approved result, verify |

## Transversal rules

- Report unsupported hosts with alternatives; a lone AGENTS file identifies no installed host.
- Stop and name any installed reference that cannot be read.
- Change existing sources or publications only with explicit permission naming them.
- Write only after the user approves the candidate shown last. A choice, merge, edit or rename is not an approval: show the revision and wait.
- Keep or cancel leaves project bytes unchanged.

© AI-Driven Dev · MIT · ai-driven-dev/framework

Similar skills

All skills for Agent setup: memory, context, orchestration →

From the same set80

The whole set →
02-challenge 02-challenge is a Claude Code skill that checks completed work against an agreed plan and returns a structured report. Its findings help decide whether to ship, revise, or rework the result. CX Codex +1 ★ 571repo stars 05-spike 05-spike turns one blocking unknown into a bounded spike: frames the question, gathers evidence, writes the outcome. CX Codex +1 ★ 571repo stars 0-contexts 0-contexts enforces one-way imports across bounded contexts and keeps business logic out of the shared kernel in TypeScript projects. CX Codex +1 ★ 571repo stars 1-conventions The 1-conventions rule sets conventions for TypeScript files under src/ and tests/, covering module structure, file names, and test tiers. Use it when adding code or tests to a project. CX Codex +1 ★ 571repo stars 07-command-generate 07-command-generate generates flat slash commands across the AI assistants enabled in your repository. Useful when you need to scaffold or edit single-step commands without manual path wiring. CX Codex +1 ★ 571repo stars 0-hexagonal The 0-hexagonal rule splits each context into four layers and keeps dependencies pointed inward. Use it when reviewing TypeScript architecture. CX Codex +1 ★ 571repo stars 09-mermaid 09-mermaid turns a written description of an architecture or a flow into a Mermaid diagram, with a plan you approve before anything gets drawn. CX Codex +1 ★ 571repo stars 06-test 06-test finds untested behaviors, iterates tests until they pass, and validates user flows end to end in the browser. CX Codex +1 ★ 571repo stars

Project sponsors

Companies that keep this analytics open