templates/skills/modules/rulebook-mcp/SKILL.md
Use MCP Rulebook server to manage tasks programmatically instead of executing terminal commands.
npx skillsauth add hivellm/rulebook Rulebook MCPInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
CRITICAL: Use MCP Rulebook server to manage tasks programmatically instead of executing terminal commands.
Create a new Rulebook task with OpenSpec-compatible format:
rulebook_task_create({
taskId: "add-feature-name",
proposal: {
why: "Users need this feature...",
whatChanges: "Add feature with X, Y, Z",
impact: {
affectedSpecs: ["specs/module/spec.md"],
affectedCode: ["src/module/"],
breakingChange: false,
userBenefit: "Better user experience"
}
}
})
List all tasks with optional filters:
rulebook_task_list({
status: "in-progress",
includeArchived: false
})
Show detailed task information:
rulebook_task_show({
taskId: "add-feature-name"
})
Update task status or progress:
rulebook_task_update({
taskId: "add-feature-name",
status: "in-progress",
progress: 50
})
Validate task format against OpenSpec requirements:
rulebook_task_validate({
taskId: "add-feature-name"
})
Archive completed task and apply spec deltas:
rulebook_task_archive({
taskId: "add-feature-name",
skipValidation: false
})
When creating tasks:
1. Use rulebook_task_create instead of terminal command
2. Provide complete proposal with why/whatChanges/impact
3. Verify task creation with rulebook_task_show
When managing task progress:
1. Use rulebook_task_list to see all tasks
2. Update status with rulebook_task_update as work progresses
3. Validate format with rulebook_task_validate before archiving
4. Archive completed tasks with rulebook_task_archive
Before archiving:
1. Always run rulebook_task_validate first
2. Fix any validation errors
3. Ensure all tasks in tasks.md are completed
4. Archive with skipValidation: false
✅ DO:
❌ DON'T:
rulebook task create commands in terminalThe Rulebook MCP server is configured in .cursor/mcp.json:
{
"mcpServers": {
"rulebook": {
"command": "node",
"args": ["dist/mcp/rulebook-server.js"],
"env": {}
}
}
}
For production (npx):
{
"mcpServers": {
"rulebook": {
"command": "npx",
"args": ["-y", "@hivehub/rulebook@latest", "mcp-server"],
"env": {}
}
}
}
The MCP server integrates seamlessly with:
.cursor/mcp.json)All task operations are available through MCP functions, eliminating the need for terminal command execution.
For complete API documentation, see:
/docs/MCP_SERVER.md - Full API reference/docs/guides/MCP_SERVER_SETUP.md - Setup guide/README.md - General project informationresearch
Author a rulebook task spec interactively — research, draft, ask the user clarifying questions, confirm, then create the tasks in rulebook ready for /rulebook-driver. Use when the user wants to plan/spec a feature before implementing.
development
Behavioral guidelines to reduce common LLM coding mistakes — overcomplication, sloppy refactors, hidden assumptions, weak goals. Use when writing, reviewing, or refactoring code. Auto-applies; invoke explicitly via /karpathy-guidelines or 'follow karpathy discipline'.
data-ai
Autonomous AI agent loop for iterative task implementation (@hivehub/rulebook ralph)
data-ai
Use SQL Server for enterprise relational data storage with advanced features, high availability, and Windows integration.