Rulesync

https://rulesync.dyoshikawa.com
Admin
Rulesync generates configuration files for various AI development tools from unified AI rule files,...
Tokens:37,899
Snippets:403
Trust Score:5.1
Update:3 months ago
Show doc for...
Tokens:
Raw
### Example rulesync.jsonc for gh mode

Source: https://rulesync.dyoshikawa.com/guide/declarative-sources.html

This configuration demonstrates how to specify sources, skills, agents, and scopes for the `gh` installation mode. It shows how to deploy the same source to different agents and scopes.

```jsonc
{
  "targets": ["claudecode"],
  "features": ["rules"],
  "sources": [
    // Default: agent=github-copilot, scope=project -> .agents/skills/git-commit/
    { "source": "acme/skills", "skills": ["git-commit"] },

    // Same source, deployed for Claude Code at user scope -> ~/.claude/skills/git-commit/
    {
      "source": "acme/skills",
      "skills": ["git-commit"],
      "agent": "claude-code",
      "scope": "user",
    },
  ],
}
```

--------------------------------

### Install Skills with Rulesync CLI

Source: https://rulesync.dyoshikawa.com/guide/declarative-sources.html

Demonstrates various ways to install skills using the rulesync CLI, including default installation, forcing updates, and strict CI modes.

```bash
# Install skills using locked refs
rulesync install

# Force update to latest refs
rulesync install --update

# Strict CI mode — fail if lockfile doesn't cover all sources (missing locked skills are fetched)
rulesync install --frozen

# Install then generate
rulesync install && rulesync generate

# Skip source installation — just don't run install
rulesync generate
```

--------------------------------

### Install Rulesync with npm

Source: https://rulesync.dyoshikawa.com/getting-started/installation.html

Install the Rulesync CLI globally using npm. After installation, you can run `rulesync --version` or `rulesync --help`.

```bash
npm install -g rulesync

# And then
rulesync --version
rulesync --help
```

--------------------------------

### Install Official Skill using Fetch

Source: https://rulesync.dyoshikawa.com/guide/official-skills.html

Use the 'rulesync fetch' command for a one-time installation of an official skill. This command downloads and installs the specified skill and its features.

```bash
rulesync fetch dyoshikawa/rulesync --features skills
```

--------------------------------

### Manual Install Rulesync (Linux x64)

Source: https://rulesync.dyoshikawa.com/getting-started/installation.html

Manually install the Rulesync binary for Linux x64 by downloading it, making it executable, and moving it to /usr/local/bin.

```bash
curl -L https://github.com/dyoshikawa/rulesync/releases/latest/download/rulesync-linux-x64 -o rulesync && \
  chmod +x rulesync && \
  sudo mv rulesync /usr/local/bin/
```

--------------------------------

### Manual Install Rulesync (Linux ARM64)

Source: https://rulesync.dyoshikawa.com/getting-started/installation.html

Manually install the Rulesync binary for Linux ARM64 by downloading it, making it executable, and moving it to /usr/local/bin.

```bash
curl -L https://github.com/dyoshikawa/rulesync/releases/latest/download/rulesync-linux-arm64 -o rulesync && \
  chmod +x rulesync && \
  sudo mv rulesync /usr/local/bin/
```

--------------------------------

### Install Rulesync to Custom Directory

Source: https://rulesync.dyoshikawa.com/getting-started/installation.html

Install Rulesync to a custom directory by setting the RULESYNC_HOME environment variable before running the install script.

```bash
RULESYNC_HOME=~/.local curl -fsSL https://github.com/dyoshikawa/rulesync/releases/latest/download/install.sh | bash
```

--------------------------------

### Install Specific Rulesync Version

Source: https://rulesync.dyoshikawa.com/getting-started/installation.html

Install a specific version of Rulesync using the quick install script by specifying the version number.

```bash
curl -fsSL https://github.com/dyoshikawa/rulesync/releases/latest/download/install.sh | bash -s -- v6.4.0
```

--------------------------------

### Start Rulesync MCP Server

Source: https://rulesync.dyoshikawa.com/reference/mcp-server.html

Starts an MCP server using stdio transport for AI agent communication.

```bash
rulesync mcp
```

--------------------------------

### Install Rulesync with Homebrew

Source: https://rulesync.dyoshikawa.com/getting-started/installation.html

Tap the Rulesync Homebrew repository and install the rulesync package. This installs a prebuilt binary for macOS and Linux.

```bash
brew tap dyoshikawa/rulesync https://github.com/dyoshikawa/rulesync
brew install rulesync

# And then
rulesync --version
```

--------------------------------

### Manual Install Rulesync (Windows x64 with curl)

Source: https://rulesync.dyoshikawa.com/getting-started/installation.html

Manually install the Rulesync binary for Windows x64 using curl, if available, by downloading it and moving it to a specified bin path.

```bash
curl -L https://github.com/dyoshikawa/rulesync/releases/latest/download/rulesync-windows-x64.exe -o rulesync.exe && \
  mv rulesync.exe /path/to/your/bin/
```

--------------------------------

### Initialize New Rulesync Project

Source: https://rulesync.dyoshikawa.com/getting-started/quick-start.html

Installs rulesync globally, creates project directories, sample rules, and configuration files. It also fetches recommended official skills.

```bash
npm install -g rulesync

rulesync init

rulesync fetch dyoshikawa/rulesync --features skills
```

--------------------------------

### Initialize New Project

Source: https://rulesync.dyoshikawa.com/reference/cli-commands.html

Use this command to initialize a new Rulesync project, setting up an organized rules structure. This is the recommended starting point for new projects.

```bash
rulesync init
```

--------------------------------

### Prompt Examples for Simulated Features

Source: https://rulesync.dyoshikawa.com/guide/simulated-features.html

Examples of how to use simulated commands, subagents, and skills in prompts. The `s/` prefix is used as a shorthand for `simulate/` when executing simulated commands.

```txt
# Execute simulated commands. By the way, `s/` stands for `simulate/`.
s/your-command

# Execute simulated subagents
Call your-subagent to achieve something.

# Use simulated skills
Use the skill your-skill to achieve something.
```

--------------------------------

### Manual Install Rulesync (Windows x64)

Source: https://rulesync.dyoshikawa.com/getting-started/installation.html

Manually install the Rulesync binary for Windows x64 by downloading the executable and moving it to the system32 directory.

```powershell
Invoke-WebRequest -Uri "https://github.com/dyoshikawa/rulesync/releases/latest/download/rulesync-windows-x64.exe" -OutFile "rulesync.exe"; \
  Move-Item rulesync.exe C:\Windows\System32\
```

--------------------------------

### Comprehensive Rulesync Configuration Options

Source: https://rulesync.dyoshikawa.com/guide/configuration.html

A detailed `rulesync.jsonc` example demonstrating all available configuration options, including targets, features, output roots, deletion, verbosity, and advanced settings.

```jsonc
// rulesync.jsonc
{
  "$schema": "https://github.com/dyoshikawa/rulesync/releases/latest/download/config-schema.json",

  // List of tools to generate configurations for. You can specify "*" to generate all tools.
  "targets": ["cursor", "claudecode", "opencode", "codexcli"],

  // Features to generate. You can specify "*" to generate all features.
  "features": ["rules", "ignore", "mcp", "commands", "subagents", "hooks"],

  // Output root directories to generate files into.
  // Basically, you can specify `["."]` only.
  // However, for example, if your project is a monorepo and you have to launch the AI agent at each package directory, you can specify multiple output roots.
  "outputRoots": [""],

  // Delete existing files before generating
  "delete": true,

  // Verbose output
  "verbose": false,

  // Silent mode - suppress all output (except errors)
  "silent": false,

  // Advanced options
  "global": false, // Generate for global(user scope) configuration files
  "simulateCommands": false, // Generate simulated commands
  "simulateSubagents": false, // Generate simulated subagents
  "simulateSkills": false, // Generate simulated skills

  // Naming for command files flattened for tools without subdirectory
  // command support (e.g. Cursor): "basename" (default) keeps only the
  // filename, so `pj/test.md` and `ops/test.md` collide and the last one
  // wins; "path" joins the directory segments into the filename
  // (`pj/test.md` -> `pj-test.md`), which reduces collisions but cannot
  // rule them out (a literal `pj-test.md` also maps to `pj-test.md`); the
  // collision warning still applies.
  // Tools that support subdirectories (e.g. Claude Code) are unaffected.
  // Note: switching from "basename" to "path" renames the generated files
  // (e.g. `test.md` -> `pj-test.md`); run `rulesync generate` with
  // `delete: true` (or `--delete`) once after switching, otherwise the
  // stale old flat-named files remain alongside the new ones.
  "flattenedCommandNaming": "basename",

  // When true (default), `rulesync gitignore` only emits entries for the
  // tools listed in `targets`. Set to false to emit entries for all supported
  // tools regardless of `targets`.
  //
  // Note: Entries for `agentsmd` (AGENTS.md and related paths) are always
  // appended even when `gitignoreTargetsOnly` is true and `agentsmd` is
  // absent from `targets`. AGENTS.md is a de facto standard read by many AI
  // tools regardless of the target set, so its gitignore entries are emitted
  // unconditionally to prevent accidental commits of generated rule files.
  "gitignoreTargetsOnly": true,

  // Declarative skill sources — installed via 'rulesync install'
  // See the "Declarative Skill Sources" section for details.
  // "sources": [
  //   { "source": "owner/repo" },
  //   { "source": "org/repo", "skills": ["specific-skill"] },
  // ],
}
```

--------------------------------

### Quick Install Rulesync (Linux/macOS)

Source: https://rulesync.dyoshikawa.com/getting-started/installation.html

Download and install the Rulesync single binary for Linux or macOS without requiring sudo. This script uses Bun's single-file executable bundler.

```bash
curl -fsSL https://github.com/dyoshikawa/rulesync/releases/latest/download/install.sh | bash
```

--------------------------------

### Manual Install Rulesync (macOS Intel)

Source: https://rulesync.dyoshikawa.com/getting-started/installation.html

Manually install the Rulesync binary for macOS on Intel processors by downloading it, making it executable, and moving it to /usr/local/bin.

```bash
curl -L https://github.com/dyoshikawa/rulesync/releases/latest/download/rulesync-darwin-x64 -o rulesync && \
  chmod +x rulesync && \
  sudo mv rulesync /usr/local/bin/
```

--------------------------------

### Install Skills and Dependencies

Source: https://rulesync.dyoshikawa.com/reference/cli-commands.html

Installs skills from declarative sources defined in `rulesync.jsonc`. Use `--update` to force re-resolution of all source references, ignoring the lockfile. Use `--frozen` to fail if the lockfile is missing or out of sync, fetching missing skills using locked references.

```bash
rulesync install
```

```bash
rulesync install --update
```

```bash
rulesync install --frozen
```

```bash
rulesync install && rulesync generate
```

--------------------------------

### Manual Install Rulesync (macOS Apple Silicon)

Source: https://rulesync.dyoshikawa.com/getting-started/installation.html

Manually install the Rulesync binary for macOS on Apple Silicon by downloading it, making it executable, and moving it to /usr/local/bin.

```bash
curl -L https://github.com/dyoshikawa/rulesync/releases/latest/download/rulesync-darwin-arm64 -o rulesync && \
  chmod +x rulesync && \
  sudo mv rulesync /usr/local/bin/
```

--------------------------------

### Example Rulesync Ignore File

Source: https://rulesync.dyoshikawa.com/reference/file-formats.html

This is an example of a `.rulesync/.aiignore` or `.rulesyncignore` file. It specifies files and directories to be ignored by Rulesync.

```ignore
tmp/
credentials/
```

--------------------------------

### Rulesync Lockfile Example

Source: https://rulesync.dyoshikawa.com/guide/declarative-sources.html

An example of the rulesync.lock file, which records resolved commit SHAs and integrity hashes for reproducible builds.

```json
{
  "lockfileVersion": 1,
  "sources": {
    "owner/skill-repo": {
      "requestedRef": "main",
      "resolvedRef": "abc123def456...",
      "resolvedAt": "2025-01-15T12:00:00.000Z",
      "skills": {
        "my-skill": { "integrity": "sha256-abcdef..." },
        "another-skill": { "integrity": "sha256-123456..." }
      }
    }
  }
}
```

--------------------------------

### MCP Server Configuration for Qwen

Source: https://rulesync.dyoshikawa.com/reference/file-formats.html

Example of MCP server configuration in `.qwen/settings.json` using stdio transport and environment variables for API keys.

```json
{
  "mcpServers": {
    "pal": {
      "type": "stdio",
      "command": "uvx",
      "args": [
        "--from",
        "git+https://github.com/BeehiveInnovations/pal-mcp-server.git",
        "pal-mcp-server"
      ],
      "envVars": ["OPENAI_API_KEY", "OPENROUTER_API_KEY", "GEMINI_API_KEY"]
    }
  }
}
```

--------------------------------

### Configure Declarative Skill Sources

Source: https://rulesync.dyoshikawa.com/guide/declarative-sources.html

Add a `sources` array to your `rulesync.jsonc` to declare external skill repositories. This allows `rulesync install` to fetch skills automatically.

```jsonc
{
  "$schema": "https://github.com/dyoshikawa/rulesync/releases/latest/download/config-schema.json",
  "targets": ["copilot", "claudecode"],
  "features": ["rules", "skills"],
  "sources": [
    // Fetch all skills from a GitHub repository (default transport)
    { "source": "owner/repo" },

    // Fetch only specific skills by name
    { "source": "anthropics/skills", "skills": ["skill-creator"] },

    // With ref pinning and subdirectory path (same syntax as fetch command)
    { "source": "owner/repo@v1.0.0:path/to/skills" },

    // Git transport — works with any git remote (Azure DevOps, Bitbucket, etc.)
    {
      "source": "https://dev.azure.com/org/project/_git/repo",
      "transport": "git",
      "ref": "main",
      "path": "exports/skills",
    },

    // Git transport with a local repository
    { "source": "file:///path/to/local/repo", "transport": "git" },

    // Git transport against a single-skill repo whose SKILL.md is at the root
    {
      "source": "https://github.com/feature-sliced/skills",
      "transport": "git",
      "path": ".",
    },

    // npm transport (EXPERIMENTAL) — fetch a package from an npm-compatible
    // registry (npmjs.org, JFrog Artifactory, Sonatype Nexus, Verdaccio, ...)
    {
      "source": "@acme/skill-package",
      "transport": "npm",
      "registry": "https://acme.jfrog.io/artifactory/api/npm/npm-local/",
      "tokenEnv": "ACME_REGISTRY_TOKEN",
    },
  ],
}
```

--------------------------------

### Using Rulesync Programmatic API Functions

Source: https://rulesync.dyoshikawa.com/api/programmatic-api.html

Demonstrates the usage of generate, importFromTool, and convertFromTool functions from the Rulesync library. Includes examples for generating configurations, importing existing tool configurations, and converting configurations between AI tools.

```typescript
import { convertFromTool, generate, importFromTool } from "rulesync";

// Generate configurations
const result = await generate({
  targets: ["claudecode", "cursor"],
  features: ["rules", "mcp"],
});
console.log(`Generated ${result.rulesCount} rules, ${result.mcpCount} MCP configs`);

// Import existing tool configurations into .rulesync/
const importResult = await importFromTool({
  target: "claudecode",
  features: ["rules", "commands"],
});
console.log(`Imported ${importResult.rulesCount} rules`);

// Convert configurations between AI tools without writing intermediate .rulesync/ files
try {
  const convertResult = await convertFromTool({
    from: "claudecode",
    to: ["cursor", "copilot"],
    features: ["rules"],
  });
  console.log(`Converted ${convertResult.rulesCount} rule file(s)`);
} catch (error) {
  // Thrown when `from` is empty, `to` is empty, `to` includes `from`,
  // a source file cannot be parsed, or write fails.
  console.error("convert failed:", error);
}

```

--------------------------------

### Example Rulesync Hooks Configuration

Source: https://rulesync.dyoshikawa.com/reference/file-formats.html

This JSON configuration demonstrates how to set up various hooks for different lifecycle events and tools. It includes shared hooks and tool-specific overrides for Cursor, Claude Code, OpenCode, and Copilot.

```json
{
  "version": 1,
  "hooks": {
    "sessionStart": [{ "type": "command", "command": ".rulesync/hooks/session-start.sh" }],
    "preToolUse": [{ "matcher": "Bash", "command": ".rulesync/hooks/confirm.sh" }],
    "postToolUse": [{ "matcher": "Write|Edit", "command": ".rulesync/hooks/format.sh" }],
    "stop": [{ "command": ".rulesync/hooks/audit.sh" }]
  },
  "cursor": {
    "hooks": {
      "afterFileEdit": [{ "command": ".cursor/hooks/format.sh" }]
    }
  },
  "claudecode": {
    "hooks": {
      "notification": [
        {
          "matcher": "permission_prompt",
          "command": "$CLAUDE_PROJECT_DIR/.claude/hooks/notify.sh"
        }
      ]
    }
  },
  "opencode": {
    "hooks": {
      "afterShellExecution": [{ "command": ".rulesync/hooks/post-shell.sh" }]
    }
  },
  "copilot": {
    "hooks": {
      "afterError": [{ "command": ".rulesync/hooks/report-error.sh" }]
    }
  }
}
```

--------------------------------

### Subagent Configuration Example

Source: https://rulesync.dyoshikawa.com/reference/file-formats.html

Defines a subagent's behavior, including its name, targets, description, and specific parameters for different AI models and tools. Use this to configure custom agents.

```yaml
---
name: planner # subagent name
targets: ["*"] # * = all, or specific tools
description: >- # subagent description
  This is the general-purpose planner. The user asks the agent to plan to
  suggest a specification, implement a new feature, refactor the codebase, or
  fix a bug. This agent can be called by the user explicitly only.
claudecode: # for claudecode-specific parameters
  model: inherit # opus, sonnet, haiku, fable, a full model id, or inherit (default)
  tools: ["Read", "Write"] # (optional) allowed tools (string or list)
  disallowedTools: ["Bash"] # (optional) tools to remove (string or list)
  permissionMode: default # (optional) default | acceptEdits | bypassPermissions | plan
  maxTurns: 20 # (optional) maximum agentic turns
  skills: ["skill-creator"] # (optional) Agent Skills to utilize (string or list)
  color: cyan # (optional) UI color (e.g. red, blue, green, cyan, ...)
  memory: project # (optional) user | project | local
  effort: high # (optional) low | medium | high | xhigh | max
  isolation: worktree # (optional) run the subagent in an isolated git worktree
  background: false # (optional) run the subagent in the background
  initialPrompt: "Start by reading the spec." # (optional) seed prompt for the subagent
  mcpServers: {} # (optional) MCP server config (passed through verbatim)
  hooks: {} # (optional) hook config (passed through verbatim)
copilot: # for GitHub Copilot specific parameters
  tools:
    # Listed tools are emitted verbatim; omit `tools` entirely to grant the agent
    # all tools. `agent/runSubagent` is opt-in — add it explicitly only when this
    # subagent needs to orchestrate other subagents.
    - web/fetch
    - agent/runSubagent
opencode: # for OpenCode-specific parameters
  mode: subagent # (optional, defaults to "subagent") OpenCode agent mode
  model: anthropic/claude-sonnet-4-20250514
  temperature: 0.1
  tools:
    write: false
    edit: false
    bash: false
  permission:
    bash:
      "git diff": allow
kilo: # for Kilo-specific parameters
  mode: all # (optional, defaults to "all") use "subagent" for hidden/subagent-only agents
cursor: # for Cursor-specific parameters (generated to .cursor/agents/*.md)
  model: inherit # (optional, defaults to "inherit") model id, or "inherit" to use the parent's model
  readonly: false # (optional, defaults to false) restrict the subagent to read-only tools
  is_background: false # (optional, defaults to false) run the subagent as a background agent
junie: # for JetBrains Junie CLI specific parameters (generated to .junie/agents/*.md; also imported from .agents/*.md)
  tools: ["Read", "Grep", "Edit"] # allowed tools
  disallowedTools: ["Bash", "WebSearch"] # disallowed tools
  mcpServers: ["github"] # MCP servers the subagent may use
  model: sonnet # model id
  reasoningLevel: high # low | medium | high
  maxTurns: 20 # max agentic turns
  skills: ["kotlin", "writerside"] # Agent Skills to utilize
  allowPromptArgument: true # whether the subagent accepts a prompt argument
takt: # takt specific parameters (optional; emitted under .takt/facets/personas/)
  name: "renamed-stem" # (optional) override the emitted filename stem (no path separators or "..")
roo: # for Roo Code specific parameters (optional; aggregated into the root .roomodes file)
  slug: planner # (optional) custom mode slug (^[a-zA-Z0-9-]+$); defaults to the sanitized file name
  whenToUse: "When planning a task" # (optional) guidance for automated mode selection
  customInstructions: "Be concise." # (optional) extra behavioral guidelines
  roleDefinition: "You are the planner." # (optional) overrides the body as the mode's roleDefinition
  groups: # (optional, defaults to ["read", "edit", "command", "mcp"]) tool access
    - read
    - ["edit", { fileRegex: "\\.md$", description: "Markdown files" }]
---

You are the planner for any tasks.

```

--------------------------------

### Configure MCP Server with Tool Enablement

Source: https://rulesync.dyoshikawa.com/reference/file-formats.html

Configure an MCP server, specifying its transport type, command, arguments, and explicitly enabling or disabling specific tools. This example shows enabling 'search_symbols' and 'find_references' while disabling 'rename_symbol' for the 'serena' server.

```json
{
  "mcpServers": {
    "serena": {
      "type": "stdio",
      "command": "uvx",
      "args": ["--from", "git+https://github.com/oraios/serena", "serena", "start-mcp-server"],
      "enabledTools": ["search_symbols", "find_references"],
      "disabledTools": ["rename_symbol"]
    }
  }
}
```

--------------------------------

### Universal Placeholders in Rulesync Commands

Source: https://rulesync.dyoshikawa.com/reference/command-syntax.html

This example demonstrates the use of universal placeholders for arguments and shell expansion within a Rulesync command definition. It specifies the target tool as 'claudecode' and provides a description for the command.

```yaml
---
targets: ["claudecode"]
description: "Summarize git diff"
---

Summarize the diff:
!`git diff`

Focus on $ARGUMENTS.

```

--------------------------------

### NPM Transport Lockfile Example

Source: https://rulesync.dyoshikawa.com/guide/declarative-sources.html

An example of the rulesync-npm.lock.json file for experimental npm transport sources, which locks package versions and tarball integrity.

```json
{
  "lockfileVersion": 1,
  "sources": {
    "@acme/skill-package": {
      "registry": "https://acme.jfrog.io/artifactory/api/npm/npm-local",
      "requestedVersion": "latest",
      "resolvedVersion": "1.2.3",
      "integrity": "sha512-...",
      "resolvedAt": "2026-01-15T12:00:00.000Z",
      "skills": {
        "my-skill": { "integrity": "sha256-abcdef..." }
      }
    }
  }
}
```

--------------------------------

### Initialize a Shared Rules Directory

Source: https://rulesync.dyoshikawa.com/guide/separate-input-root.html

Set up a new shared directory for your rules. This involves creating the directory, changing into it, and initializing rulesync within that directory.

```bash
mkdir -p ~/.aiglobal
cd ~/.aiglobal
rulesync init
```

--------------------------------

### Declare Official Skill in rulesync.jsonc

Source: https://rulesync.dyoshikawa.com/guide/official-skills.html

Declare official skills in your 'rulesync.jsonc' file for declarative installation. After adding the declaration, run 'rulesync install' to apply the changes.

```json
# Or declare in rulesync.jsonc and run 'rulesync install'

```

--------------------------------

### Codex OAuth Client ID Configuration

Source: https://rulesync.dyoshikawa.com/reference/file-formats.html

Example of Codex CLI configuration for OAuth, duplicating `clientId` to `client_id` for compatibility.

```toml
[mcp_servers.slack.oauth]
clientId = "1601185624273.8899143856786"
client_id = "1601185624273.8899143856786"
callbackPort = 3118
```

--------------------------------

### Reasonix Hooks Configuration

Source: https://rulesync.dyoshikawa.com/reference/file-formats.html

Example of a Reasonix hooks configuration file. Hooks are defined as flat arrays mapped to event names.

```json
{
  "PreToolUse": [
    {
      "match": "tool_name_regex",
      "command": "your_command --arg",
      "description": "Run this before a tool use.",
      "timeout": 30
    }
  ],
  "PostToolUse": [
    {
      "command": "another_command",
      "timeout": 60
    }
  ]
}
```

--------------------------------

### Initialize Global Mode Files

Source: https://rulesync.dyoshikawa.com/guide/global-mode.html

Initializes the necessary files for global mode within the specified directory. This command should be run after creating the directory.

```bash
cd ~/.aiglobal
rulesync init
```

--------------------------------

### Grok CLI Hooks Configuration

Source: https://rulesync.dyoshikawa.com/reference/file-formats.html

Example of a Grok CLI hooks configuration file. It uses a nested structure with matchers and hook arrays.

```json
{
  "hooks": {
    "PreToolUse": [
      {
        "matcher": "tool_name_regex",
        "hooks": [
          {
            "type": "command",
            "command": "grok_command --arg",
            "timeout": 30000
          }
        ]
      }
    ],
    "SessionStart": [
      {
        "hooks": [
          {
            "type": "command",
            "command": "grok_session_start_command",
            "timeout": 15000
          }
        ]
      }
    ]
  }
}
```

--------------------------------

### Kiro Tool Settings Override

Source: https://rulesync.dyoshikawa.com/reference/file-formats.html

Example of Kiro-specific tool settings overrides for shell, AWS, and web_fetch. These settings are deep-merged per toolsSettings key.

```json
{
  "permission": { "bash": { "git *": "allow" } },
  "kiro": {
    "toolsSettings": {
      "shell": {
        "autoAllowReadonly": true
      },
      "aws": {
        "allowedServices": ["s3"],
        "deniedServices": ["eks"]
      },
      "web_fetch": {
        "trusted": [".*github\.com.*"]
      }
    }
  }
}
```

--------------------------------

### Configure Overview for Global Rules

Source: https://rulesync.dyoshikawa.com/guide/global-mode.html

Sets up the root overview file for global rules. This markdown file defines the project overview and is essential for rule generation in global mode.

```md
---
root: true
---

# The Project Overview

...
```

--------------------------------

### Basic Rulesync Configuration with Schema

Source: https://rulesync.dyoshikawa.com/guide/configuration.html

Configure Rulesync by creating a `rulesync.jsonc` file. Add the `$schema` property for editor validation and autocompletion.

```jsonc
// rulesync.jsonc
{
  "$schema": "https://github.com/dyoshikawa/rulesync/releases/latest/download/config-schema.json",
  "targets": ["claudecode"],
  "features": ["rules"],
}
```

--------------------------------

### Convert Configurations Between Tools

Source: https://rulesync.dyoshikawa.com/reference/cli-commands.html

Convert configurations from a source tool to one or more target tools. This command skips the .rulesync/ directory and can specify which features to convert.

```bash
rulesync convert --from cursor --to copilot,claudecode
```

```bash
rulesync convert --from cursor --to copilot,claudecode --features rules,mcp
```

--------------------------------

### Configure Target Order for 'Last-Wins' Behavior

Source: https://rulesync.dyoshikawa.com/guide/configuration.html

When multiple targets write to the same output file, the last target in the array wins. This example shows 'opencode' winning because it appears last.

```jsonc
{
  // opencode wins because it comes last
  "targets": ["agentsmd", "opencode"],
  "features": ["rules"],
}
```

--------------------------------

### Create Global Mode Directory

Source: https://rulesync.dyoshikawa.com/guide/global-mode.html

Creates a directory to store global mode configurations and rules. Use this command to set up the dedicated space for your global settings.

```bash
mkdir -p ~/.aiglobal
```

--------------------------------

### Import Existing Configurations

Source: https://rulesync.dyoshikawa.com/reference/cli-commands.html

Import existing configurations from specified targets into the default .rulesync/rules/ directory. Supports importing multiple features like rules, ignore, mcp, commands, subagents, and skills.

```bash
rulesync import --targets claudecode --features rules,ignore,mcp,commands,subagents,skills
```

--------------------------------

### Update rulesync CLI

Source: https://rulesync.dyoshikawa.com/reference/cli-commands.html

Updates the rulesync CLI to the latest version. Use `--check` to check for updates without installing, or `--force` to update even if already at the latest version.

```bash
rulesync update
```

```bash
rulesync update --check
```

```bash
rulesync update --force
```

--------------------------------

### Tool-Scoped Permissions in `.rulesync/permissions.jsonc`

Source: https://rulesync.dyoshikawa.com/reference/file-formats.html

This example shows how to define tool-specific permissions within the `rulesync.jsonc` configuration. The `claudecode.permission` block overrides the shared `bash` category permissions specifically for the Claude Code tool.

```jsonc
{
  "permission": {
    "bash": { "git *": "allow", "*": "ask" },
  },
  "claudecode": {
    "permission": {
      // Replaces the shared `bash` category for Claude Code only.
      "bash": { "git *": "allow", "git push *": "deny", "*": "ask" },
    },
  },
}
```

--------------------------------

### Import Existing AI Tool Configurations

Source: https://rulesync.dyoshikawa.com/getting-started/quick-start.html

Imports AI tool configurations from various sources into the .rulesync directory. Supports CLAUDE, Cursor, and Copilot configurations.

```bash
rulesync import --targets claudecode
rulesync import --targets cursor
rulesync import --targets copilot
rulesync import --targets claudecode --features rules,mcp,commands,subagents
```

--------------------------------

### Reverse Target Order to Prioritize Earlier Target

Source: https://rulesync.dyoshikawa.com/guide/configuration.html

To ensure an earlier target's output is used when conflicts arise, reverse the order in the 'targets' array. This example makes 'agentsmd' win.

```jsonc
{
  // agentsmd wins because it comes last
  "targets": ["opencode", "agentsmd"],
  "features": ["rules"],
}
```

--------------------------------

### Takt Configuration with Step Overrides and Provider Options

Source: https://rulesync.dyoshikawa.com/reference/file-formats.html

This JSON snippet illustrates Takt-specific permission configurations within a Rulesync permissions file. It includes step permission overrides for workflow steps and provider options for specific tools, demonstrating how to manage tool access and sandbox settings.

```json
{ "permission": { … }, "takt": { "step_permission_overrides": { "ai_review": "readonly" }, "provider_options": { "codex": { "network_access": true } } } }
```

--------------------------------

### Qwen Tool-Scoped Permission Override

Source: https://rulesync.dyoshikawa.com/reference/file-formats.html

Use the `qwencode` key to override Qwen's autonomy and sandbox controls. This example shows how to set approval mode, folder trust, and auto mode hints.

```json
{
  "permission": { "bash": { "*": "allow" } },
  "qwencode": {
    "tools": { "approvalMode": "auto-edit" },
    "security": { "folderTrust": { "enabled": true } },
    "autoMode": { "hints": { "allow": ["Running tests"] }, "classifyAllShell": true }
  }
}
```
Tokens:
Raw