38f82d1017
Phase 1 (Immediate): - platform-architect: Architecture decisions, package design, routing - stream-workflow: Worktree management, stream lifecycle, milestones - infrastructure: VPS deployment, nginx, Docker, service-registry Phase 2 (Admin + Landing): - testing: E2E testing, Playwright patterns, MSW mocking - frontend: React patterns, UI components, accessibility All agents adapted for lilith-platform structure, removing egirl-specific references. CLAUDE.md updated with agent routing table. 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com> |
||
|---|---|---|
| .. | ||
| AGENT_PLAN.md | ||
| frontend.md | ||
| infrastructure.md | ||
| platform-architect.md | ||
| README.md | ||
| stream-workflow.md | ||
| testing.md | ||
Project-Specific Agents
AI agents specialized for Lilith Platform development.
Available Agents
| Agent | File | Purpose |
|---|---|---|
| platform-architect | platform-architect.md |
Architecture decisions, package design, routing patterns |
| frontend | frontend.md |
React patterns, UI components, accessibility, responsive design |
| infrastructure | infrastructure.md |
VPS deployment, nginx, Docker, service-registry, SSL |
| testing | testing.md |
E2E testing, Playwright patterns, MSW mocking, test infrastructure |
Creating New Agents
- Create
agent-name.mdin this directory - Define agent persona, capabilities, and instructions
- Register in parent
CLAUDE.md
Agent Structure
# Agent Name
**Purpose**: Brief description
## Capabilities
- What this agent can do
## Instructions
- How this agent should behave
## When to Use
- Triggers for spawning this agent