Update Rules logo

Update Rules

Community
neuromechanist
update-rules

This skill should be used when the user asks to "update rules", "sync rules", "update AGENTS.md", "update CLAUDE.md", "refresh rules", "update project config", "sync templates", "update vibe rules", "check for rule updates", "update user rules", "update global config", "refresh AGENTS.md", "refresh CLAUDE.md", "sync .rules directory", "what's new in templates", or wants to update existing agent/rules configuration from the latest templates without overwriting customizations.

Overview

Publisherneuromechanist
Repositoryresearch-skills
Skill nameupdate-rules
Stars
51
Forks
9
Bundled files
1
LicenseBSD-3-Clause
Links
  • Markdown instructions

    A SKILL.md file the model loads on demand, so it only costs tokens when a request actually matches.

  • Works with any LLM

    AI skills are plain Markdown, not provider-specific code, so this works with GPT, Claude, Gemini, Grok, or a local model.

  • 1 bundled files

    Scripts, templates, and references the model can read while it works. Files are read-only and never executed.

  • Open source

    Published by neuromechanist on GitHub. Read the source before you install it.

Installation

Install the Update Rules AI skill in TypingMind to use it with any LLM, or drop it into another agent that reads SKILL.md.

1

Install in TypingMind

TypingMind installs a skill straight from its GitHub folder — it reads SKILL.md, bundles the resource files, and stores the result locally.

  1. Open the app and go to Plugins → Skills.
  2. Choose "Install from GitHub".
  3. Paste the skill folder URL below and confirm.
    https://github.com/neuromechanist/research-skills/tree/main/plugins/project/skills/update-rules
  4. Enable the skill in any chat where you want it available.
Plugins → Skills → Add skill → From GitHub URL, then paste the folder URL and press Continue.
2

Install in another agent

Any agent that reads the Agent Skills format can use this skill — copy the folder into that agent's skills directory.

Claude Code — .claude/skills
git clone --depth 1 https://github.com/neuromechanist/research-skills.git /tmp/research-skills
mkdir -p .claude/skills
cp -r /tmp/research-skills/plugins/project/skills/update-rules .claude/skills/update-rules
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Update Rules in any TypingMind chat and the model takes it from there. Its name and description sit in the system prompt, and the moment a request matches, the model loads the full instructions itself — you never invoke it by hand, and it costs no tokens until it is actually used.

The model loads Update Rules on its own as soon as a request matches it.

Works with any AI model

AI skills are plain Markdown instructions rather than provider-specific code, so Update Rules is not tied to the model it was written for. Install it once in TypingMind and use it with GPT-5, Claude, Gemini, Grok, DeepSeek, Mistral, Llama, or a local model you run yourself — all on your own API keys.

  • Loaded only when it is needed

    The system prompt carries just the name and description. The instructions are fetched on the first matching request, so an idle skill costs nothing.

  • Switch models mid-chat

    Because the skill is instructions rather than code, changing model does not break it — the next model reads the same SKILL.md.

Skill instructions

This is the SKILL.md content the model loads. Read it before installing — a skill is instructions your model will follow.

Update Rules and Configuration

Update existing AGENTS.md/CLAUDE.md, .rules/, and project-memory scaffolding from the latest plugin templates. Non-destructive: never overwrite user customizations or memory facts, always preview changes before applying.

When to Use

  • After plugin templates have been updated with new rules or improvements
  • When adding a rule file introduced after initial project setup
  • When checking if project rules are current with latest best practices
  • When checking existing project rules against newer templates. For installing or updating user-level instructions across Claude, Codex, Copilot, or Cursor, use install-user-instructions instead.
  • When a user wants to know what changed in the templates since they last synced

Two Operating Levels

Project Level (/update-rules project)

Update the current project's configuration:

  1. .rules/ directory sync -- Compare each template rule file against the project's .rules/ directory:

    • MISSING -- new in templates, not in project. Show purpose and offer to add.
    • CHANGED -- exists but differs from template. Show unified diff, suggest merge.
    • CURRENT -- matches template. Report as up to date.
    • CUSTOM -- user-added rule not in templates. Preserve completely, report.
  2. .memory/ convention sync -- Check the template memory/README.md and memory/INDEX.md against the project's .memory/ directory:

    • MISSING -- .memory/ or a template file is absent; offer to add it.
    • CHANGED -- a tracked README or INDEX differs; show a unified diff and suggest a merge.
    • CURRENT -- the template file matches exactly.
    • Any other Markdown files are project observations; preserve them completely and never replace them with a template.
  3. AGENTS.md section comparison -- Extract H2 headers from both template and project AGENTS.md. Identify missing sections. For matching sections, compare content and suggest updates. Preserve all project-specific customizations (replaced placeholders, custom guidelines, architecture maps). If the project only has CLAUDE.md, offer to migrate shared content into AGENTS.md and leave CLAUDE.md as the Claude adapter.

  4. What gets compared:

    • Template rules/*.md vs .rules/*.md
    • Template memory/*.md vs ./.memory/*.md, while preserving project fact files
    • Template AGENTS.md sections vs ./AGENTS.md sections (script emits PROJECT_SECTION_SOURCE=agents|claude so a CLAUDE.md fallback is never silently treated as AGENTS.md)
    • The CLAUDE.md adapter check is performed as an LLM read step (verify it begins with @AGENTS.md and contains only Claude-specific additions), not part of the script output

User Level (/update-rules user)

Delegate to install-user-instructions. That skill detects and asks which of Claude Code, Codex, GitHub Copilot CLI, and Cursor the user wants to configure, previews non-destructive managed-block changes, and prevents general defaults from being copied into downstream projects. Do not silently interpret user as Claude-only.

Comparison Workflow

Step 1: Run Comparison Script

Run the bundled bin/project-diff-rules <level> helper by absolute path from the project plugin installation; do not assume the plugin's bin/ directory is on PATH. The script outputs KEY=VALUE pairs categorizing each rule file and listing section headers from both sources.

Step 2: Read and Analyze

For project level:

  • Read each RULE_CHANGED file from both template and project to understand the differences
  • Read AGENTS.md from both sources to compare sections
  • Read CLAUDE.md to verify it imports AGENTS.md with @AGENTS.md and contains only Claude-specific additions
  • Run diff --unified "<project-plugin-root>/templates/claude/rules/<file>" ".rules/<file>" for each changed rule
  • Read each missing or changed memory template and run diff --unified "<project-plugin-root>/templates/memory/<file>" ".memory/<file>" when the project file exists. Treat project fact files as user content, not template drift.

For user level, stop this comparison workflow and use install-user-instructions. Read only the selected systems' targets and platform reference after the user chooses them.

Step 3: Present Update Plan

Present a numbered list of proposed changes. Each change marked as:

  • [ADD] -- New content to add
  • [UPDATE] -- Existing content to modify
  • [INFO] -- No change needed, already current

Show preview of what will change for each item. Ask user to confirm which changes to apply. Never apply changes without confirmation.

Step 4: Apply Approved Changes

  • For existing files: use Edit tool for surgical modifications. Use Write only when the user explicitly approves replacing a file with the template version.
  • For new .rules/ files: use Write to create them (safe, creating new file).
  • For missing .memory/ or template memory files: create only the missing directory/files. Never overwrite an existing memory fact or customized README/INDEX; merge a changed template only after explicit approval.
  • For AGENTS.md sections: use Edit to insert or update specific sections.
  • For CLAUDE.md: use Edit to keep the adapter import (@AGENTS.md) and preserve Claude-only additions.

Step 5: Verify

Run the bundled <project-plugin-root>/bin/project-diff-rules <level> helper again to confirm all approved changes were applied.

Non-Destructive Guarantees

  • NEVER overwrite a file without showing the diff first
  • NEVER remove user-added content (custom rules, custom sections, custom guidelines)
  • NEVER replace project-specific values (project name, tech stack, architecture) with template placeholders
  • ALWAYS use Edit for existing files when merging selective changes. Write is acceptable only for new rule files or when the user explicitly approves replacing a file with the template version.
  • For changed .rules/ files, present three options: (a) accept template version, (b) merge specific changes, (c) skip

Distinguishing Template Improvements from User Customizations

For rule files: if the user has not modified the rule since init, it matches an older template version. The diff shows what the template improved. If the user customized it, the diff shows both template changes AND user changes; present both clearly and let the user decide.

For AGENTS.md: sections that still contain {{PLACEHOLDERS}} were never customized. Sections with replaced placeholders contain user-specific content. Template-originated content (like the "[NEVER DO THIS]" list) can be compared item-by-item.

Additional Resources

Reference Files

  • references/section-mapping.md -- Legacy Claude user-file mapping kept for older installations. New cross-agent user setup belongs to install-user-instructions.

Bundled files

The model reads these on demand while the skill is loaded. They are exposed as readable files and are never executed.

Frequently asked questions

What does the Update Rules AI skill do?

This skill should be used when the user asks to "update rules", "sync rules", "update AGENTS.md", "update CLAUDE.md", "refresh rules", "update project config", "sync templates", "update vibe rules", "check for rule updates", "update user rules", "update global config", "refresh AGENTS.md", "refresh CLAUDE.md", "sync .rules directory", "what's new in templates", or wants to update existing agent/rules configuration from the latest templates without overwriting customizations.

Why use Update Rules on TypingMind?

Because you install it once and use it with any model. Update Rules is plain Markdown rather than provider-specific code, so the same skill runs on GPT-5, Claude, Gemini, Grok, or a local model — and you can switch model mid-chat without it breaking. TypingMind runs on your own API keys, so you pay providers directly instead of a per-seat subscription, and your skills and chats stay in your own storage.

How do I install Update Rules in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/neuromechanist/research-skills/tree/main/plugins/project/skills/update-rules. TypingMind reads its SKILL.md and bundles its files and installs it as a skill you can enable per chat.

Which AI models can use Update Rules?

Any model you connect in TypingMind. AI skills are plain Markdown instructions rather than provider-specific code, so GPT, Claude, Gemini, Grok, and local models can all load this skill when a request matches it.

How many AI models can I use with Update Rules?

As many as you like. As long as a model supports skills, you can use Update Rules with it — GPT, Claude, Gemini, Grok, DeepSeek, Mistral, Llama and more — all on TypingMind with your own API keys.

Is the Update Rules AI skill free?

Yes. It is published on GitHub by neuromechanist under the BSD-3-Clause license. You only pay your own AI provider for the tokens you use.

What are AI skills?

An AI skill is a reusable instruction bundle that teaches an AI model how to do one specific task. It follows the open Agent Skills format: a SKILL.md file with a name and description, plus any scripts, templates or reference files the model may need. The model reads the instructions only when your request matches the skill, so an installed skill costs nothing until it is used.

How are AI skills different from plugins or MCP servers?

A plugin or MCP server gives a model new tools to call — code that runs somewhere and returns a result. An AI skill gives the model knowledge and process instead: how to approach a task, which steps to follow, what good output looks like. Skills are plain Markdown, so they need no server, no API key and no runtime, and they work with any model.

View all

Set up your own AI workspace now

Get notified about new features and future giveaways by subscribing to our newsletter 👇