domain-router

Routes a keyword or problem to the correct repos, components, and documentation scope. Part of the agent-grounding stack.

Problem

Agents often jump to random logs, processes, or services without first clarifying:

• Which system is actually meant?
• Which repos are relevant?
• Which documents should be read first?

Install

npm install -g @lannguyensi/domain-router

Usage

# Route a keyword to its scope
domain-router route -k clawd-monitor -w /projects

# JSON output for scripting
domain-router route -k clawd-monitor -w /projects --json

# Show which repos depend on a keyword/package (reverse lookup)
domain-router impact -k clawd-monitor -w /projects

# JSON output for scripting
domain-router impact -k clawd-monitor -w /projects --json

Example Output

🗂  Domain Router — "clawd-monitor"

  Domain:      clawd-monitor
  Confidence:  92%

  Primary Repos:
    📁 clawd-monitor
    📁 clawd-monitor-agent

  Related Components:
    - Web UI
    - Backend API
    - OpenClaw Gateway

  Priority Files (read first):
    📄 README.md
    📄 AGENT_ENTRYPOINT.yaml
    📄 .env.example

  ❌ Forbidden initial jumps:
    - random log search
    - unrelated service inspection
    - network diagnosis before process check

API

import { route } from '@lannguyensi/domain-router';

const result = route({
  keyword: 'clawd-monitor',
  workspace: '/projects',
  context: { host: 'vps-01', problem_hint: 'agent not visible' }
});
// → { domain, primary_repos, related_components, priority_files, forbidden_initial_jumps, confidence }

Part of the grounding stack

1. domain-router ← you are here
2. readme-first-resolver
3. debug-playbook-engine
4. evidence-ledger