Stats
Actions
Tags
From aura-frog
Directory of 10 specialized subagents for delegating tasks in system architecture, frontend/mobile development, business strategy, testing, security audits, DevOps pipelines, and workflow coordination. Auto-selects agents by task keywords, tech stack, and files.
How this agent operates — its isolation, permissions, and tool access model
Agent reference
aura-frog:agents/readmeThe summary Claude sees when deciding whether to delegate to this agent
**Format:** [TOON](https://github.com/toon-format/toon) (Token-Optimized) **Last Updated:** 2026-02-09 --- Aura Frog provides **10 specialized agents** (reduced from 15 in v1.17.0 through consolidation, voice-operations removed in v1.18.1, strategist added in v2.0.0, gamedev externalized in v1.23.0). --- | New Agent | Replaces | Purpose | |-----------|----------|---------| | **scanner** | proje...
Format: TOON (Token-Optimized) Last Updated: 2026-02-09
Aura Frog provides 10 specialized agents (reduced from 15 in v1.17.0 through consolidation, voice-operations removed in v1.18.1, strategist added in v2.0.0, gamedev externalized in v1.23.0).
| New Agent | Replaces | Purpose |
|---|---|---|
| scanner | project-detector, project-config-loader, project-context-manager | Unified project detection, config, and context |
| architect | backend-expert, database-specialist | System design + database architecture |
| frontend | web-expert, ui-designer | Frontend + design systems |
agents[10]{name,file,expertise}:
architect,architect.md,System design + database + backend architecture
frontend,frontend.md,Frontend frameworks + design systems + accessibility
mobile,mobile.md,React Native + Flutter + mobile platforms
strategist,strategist.md,Business strategy + product thinking + ROI evaluation
tester,tester.md,Testing strategies + QA + TDD enforcement
security,security.md,Security audits + vulnerability assessment
devops,devops.md,CI/CD pipelines + deployment + infrastructure
lead,lead.md,Workflow coordination + team lead
scanner,scanner.md,Project detection + config + context
router,router.md,Agent selection + routing logic
Agents are automatically selected by the agent-detector skill based on:
Manual activation:
agent:activate <agent-name>
List all agents:
agent:list
| Category | Agents | Purpose |
|---|---|---|
| Development | architect, frontend, mobile | Code implementation |
| Strategy | strategist | Business thinking + product decisions |
| Quality | tester, security | Testing + security |
| Infrastructure | devops | Deployment + CI/CD |
| Management | lead | Workflow coordination |
| System | scanner, router | Internal operations |
| Version | Agents | Change |
|---|---|---|
| v1.16.0 | 15 | - |
| v1.17.0 | 11 | -4 (consolidated) |
| v1.18.1 | 10 | -1 (voice-operations removed) |
| v2.0.0 | 11 | +1 (strategist added) |
| v1.23.0 | 10 | -1 (gamedev externalized as addon) |
Removed files (consolidated):
skills/agent-detector/SKILL.mdskills/lazy-agent-loader/SKILL.mddocs/AGENT_SELECTION_GUIDE.mddocs/REFACTOR_ANALYSIS.mdnpx claudepluginhub nguyenthienthanh/aura-frog --plugin aura-frogA collection of 72 specialized agent personas for Claude Code, each with domain-specific capabilities, tools, and expertise for tasks like security auditing, DevOps, ML ops, compliance validation, and R development.
Creates, improves, and analyzes Claude Code agents following Anthropic best practices. Delegate for explicit requests like 'create agent', 'new agent', 'improve agent', 'fix agent trigger', or 'agent doesn't trigger'.
Intelligent agent router that selects primary, secondary agents via multi-layer analysis of task content, natural language intent, project context, file patterns, and domain matching with confidence scoring.