profile-selector

Match task keywords to profiles and return configuration with confidence and thresholds.

Updated Apr 14, 2026
One-click install
npx skills add https://github.com/bianhaifeng789-hue/openclaw-config --skill profile-selector
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: profile-selector
Source: https://github.com/bianhaifeng789-hue/openclaw-config/tree/main/skills/profile-selector
Command: npx skills add https://github.com/bianhaifeng789-hue/openclaw-config --skill profile-selector

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Automatically maps a user's task description to the most appropriate operational profile so workflows run with the right budget, middleware, and pass thresholds; this reduces manual configuration and speeds task routing and execution.

Core Features & Use Cases

  • Priority-based profile matching: Uses ordered priorities (app-builder, swe-bench, reasoning, terminal) to resolve conflicts when multiple profiles match.
  • Keyword-driven routing: Matches task keywords to profiles (terminal, app-builder, swe-bench, reasoning) and enforces profile-specific budgets, thresholds, and middleware.
  • Consistent output contract: Returns a structured result including profile name, confidence, matchedKeyword, taskBudget, passThreshold, and maxRounds for downstream orchestration.
  • Use case: Given a user prompt like "Build a Pomodoro timer", the selector will identify app-builder with high confidence and provide the configuration to run the build workflow.

Quick Start

Detect the appropriate profile and return its configuration for the task "Build a Pomodoro timer".

Frequently Asked Questions about profile-selector

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I automatically select the optimal configuration profile for a CLI task?

Profile matching selects the optimal configuration profile for a CLI task by matching task keywords to profile definitions and returning the chosen profile with its specific budget and thresholds.

How does keyword-driven task routing work for web app builds?

Task routing for web app builds works by matching task keywords to profile definitions like app-builder, resolving conflicts using ordered priorities, and returning the matched profile and confidence level.

What is the priority order when multiple profiles match a reasoning query?

When multiple profiles match a reasoning query, the priority order app-builder, swe-bench, reasoning, terminal is applied to resolve conflicts and select the highest priority profile.

Can I use keyword matching to set task budget and pass thresholds for GitHub issue fixes?

Yes, keyword matching for GitHub issue fixes applies the swe-bench profile, returning a structured output that includes taskBudget, passThreshold, and maxRounds for downstream orchestration.

What configuration details are returned when task routing identifies a terminal profile?

Task routing for a terminal profile returns a structured output contract containing the profile name, confidence level, matchedKeyword, taskBudget, passThreshold, and maxRounds.