- Home
- Skills
- Vudovn
- Antigravity Kit
- Intelligent Routing
intelligent-routing_skill
- TypeScript
5.7k
GitHub Stars
1
Bundled Files
2 months ago
Catalog Refreshed
4 months ago
First Indexed
Readme & install
Copy the install command, review bundled files from the catalogue, and read any extended description pulled from the listing source.
Installation
Preview and clipboard use veilstrat where the catalogue uses aiagentskills.
npx veilstrat add skill vudovn/antigravity-kit --skill intelligent-routing- SKILL.md10.4 KB
Overview
This skill provides automatic agent selection and intelligent task routing. It analyzes user requests and transparently applies the best specialist agent(s) so users get expert responses without needing to name agents or use commands. It falls back to an orchestrator for multi-domain or complex tasks and allows explicit agent overrides.
How this skill works
Every incoming request is silently analyzed for keywords, domains, and complexity. A selection matrix maps detected patterns to one or more specialist agents and auto-invokes them for simple or moderate tasks. If two or more domains or high complexity are detected, the skill routes the request to an orchestrator that coordinates multiple agents and asks clarifying questions when needed.
When to use it
- Routing single-domain requests to the right specialist automatically (e.g., frontend, backend, security).
- Coordinating multi-domain work by invoking an orchestrator for architecture or cross-cutting features.
- Reducing back-and-forth when users don’t know which specialist to ask.
- Applying transparent expertise in responses by indicating which agent(s) were used.
- Allowing users to explicitly override routing by mentioning an agent.
Best practices
- Keep analysis silent; only inform users which agent(s) are being applied, not the analysis steps.
- Auto-invoke single specialists for clear, single-domain tasks to minimize latency.
- Use the orchestrator for tasks affecting multiple domains or requiring architecture decisions.
- Ask clarifying (Socratic) questions when requirements are vague before routing.
- Allow explicit agent mentions to override automatic selection when users prefer control.
Example use cases
- User: 'Fix the login 401 error' → Auto-invoke debugger and security-auditor for a security-focused debug.
- User: 'Create a dark mode toggle button' → Auto-invoke frontend-specialist and return a UI-focused implementation.
- User: 'Add user profile API endpoint and DB schema' → Auto-invoke backend-specialist + database-architect.
- User: 'Build a chat app with real-time notifications' → Route to orchestrator to coordinate backend, frontend, and testing.
- User: 'How does React work?' → Treat as a general question and respond directly without specialist routing.
FAQ
Yes. Users can explicitly mention an agent (e.g., @backend-specialist) to override automatic selection.
What happens for vague or contradictory requests?
The skill asks clarifying questions first (Socratic Gate) and then routes to the appropriate agent(s).